quantum information theory...
TRANSCRIPT
![Page 1: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/1.jpg)
Quantum Information Theory Tutorial
Mark M. Wilde
Hearne Institute for Theoretical Physics,Department of Physics and Astronomy,
Center for Computation and Technology,Louisiana State University,
Baton Rouge, Louisiana, USA
Reference: Quantum Information Theorypublished by Cambridge University Press (2nd edition forthcoming)
July 10, 2016, ISIT 2016, Barcelona, Spain
Mark M. Wilde (LSU) 1 / 113
![Page 2: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/2.jpg)
Main questions
What are the ultimate limitations on communication imposed byphysical laws?
What are methods for achieving these limits?
To address these questions, we need to consider quantum mechanics,and so we are naturally led to an intersection of information theoryand quantum mechanics called quantum information theory
What is different about quantum and “classical” information theory?
What tasks can we achieve with quantum mechanics that we cannotwithout it? (long list: Bell inequalities, super-dense coding,teleportation, data locking, data hiding, quantum cryptography, etc.)
Mark M. Wilde (LSU) 2 / 113
![Page 3: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/3.jpg)
Main questions
What are the ultimate limitations on communication imposed byphysical laws?
What are methods for achieving these limits?
To address these questions, we need to consider quantum mechanics,and so we are naturally led to an intersection of information theoryand quantum mechanics called quantum information theory
What is different about quantum and “classical” information theory?
What tasks can we achieve with quantum mechanics that we cannotwithout it? (long list: Bell inequalities, super-dense coding,teleportation, data locking, data hiding, quantum cryptography, etc.)
Mark M. Wilde (LSU) 2 / 113
![Page 4: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/4.jpg)
Main questions
What are the ultimate limitations on communication imposed byphysical laws?
What are methods for achieving these limits?
To address these questions, we need to consider quantum mechanics,and so we are naturally led to an intersection of information theoryand quantum mechanics called quantum information theory
What is different about quantum and “classical” information theory?
What tasks can we achieve with quantum mechanics that we cannotwithout it? (long list: Bell inequalities, super-dense coding,teleportation, data locking, data hiding, quantum cryptography, etc.)
Mark M. Wilde (LSU) 2 / 113
![Page 5: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/5.jpg)
Main questions
What are the ultimate limitations on communication imposed byphysical laws?
What are methods for achieving these limits?
To address these questions, we need to consider quantum mechanics,and so we are naturally led to an intersection of information theoryand quantum mechanics called quantum information theory
What is different about quantum and “classical” information theory?
What tasks can we achieve with quantum mechanics that we cannotwithout it? (long list: Bell inequalities, super-dense coding,teleportation, data locking, data hiding, quantum cryptography, etc.)
Mark M. Wilde (LSU) 2 / 113
![Page 6: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/6.jpg)
Main questions
What are the ultimate limitations on communication imposed byphysical laws?
What are methods for achieving these limits?
To address these questions, we need to consider quantum mechanics,and so we are naturally led to an intersection of information theoryand quantum mechanics called quantum information theory
What is different about quantum and “classical” information theory?
What tasks can we achieve with quantum mechanics that we cannotwithout it? (long list: Bell inequalities, super-dense coding,teleportation, data locking, data hiding, quantum cryptography, etc.)
Mark M. Wilde (LSU) 2 / 113
![Page 7: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/7.jpg)
Main questions
What are the ultimate limitations on communication imposed byphysical laws?
What are methods for achieving these limits?
To address these questions, we need to consider quantum mechanics,and so we are naturally led to an intersection of information theoryand quantum mechanics called quantum information theory
What is different about quantum and “classical” information theory?
What tasks can we achieve with quantum mechanics that we cannotwithout it? (long list: Bell inequalities, super-dense coding,teleportation, data locking, data hiding, quantum cryptography, etc.)
Mark M. Wilde (LSU) 2 / 113
![Page 8: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/8.jpg)
Prehistory of quantum information theory
1927 Heisenberg uncertainty principle
1935 Einstein–Podolsky–Rosen paper questioning compatibility ofuncertainty principle and phenomenon of quantum entanglement /1964 Bell’s theorem as an answer / 2009 Berta et al. entropicuncertainty relation as another answer
1932 von Neumann quantum entropy / 1962 Umegaki quantumrelative entropy / 1973 Lieb–Ruskai strong subadditivity of quantumentropy / 1975 Lindblad data-processing for quantum relative entropy
1970s theory of quantum measurements and similarity measures forquantum states — Helstrom, Holevo (Shannon Award 2016), Ozawa,Bures, Uhlmann, etc.
Mark M. Wilde (LSU) 3 / 113
![Page 9: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/9.jpg)
Prehistory of quantum information theory
1927 Heisenberg uncertainty principle
1935 Einstein–Podolsky–Rosen paper questioning compatibility ofuncertainty principle and phenomenon of quantum entanglement /1964 Bell’s theorem as an answer / 2009 Berta et al. entropicuncertainty relation as another answer
1932 von Neumann quantum entropy / 1962 Umegaki quantumrelative entropy / 1973 Lieb–Ruskai strong subadditivity of quantumentropy / 1975 Lindblad data-processing for quantum relative entropy
1970s theory of quantum measurements and similarity measures forquantum states — Helstrom, Holevo (Shannon Award 2016), Ozawa,Bures, Uhlmann, etc.
Mark M. Wilde (LSU) 3 / 113
![Page 10: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/10.jpg)
Prehistory of quantum information theory
1927 Heisenberg uncertainty principle
1935 Einstein–Podolsky–Rosen paper questioning compatibility ofuncertainty principle and phenomenon of quantum entanglement /1964 Bell’s theorem as an answer / 2009 Berta et al. entropicuncertainty relation as another answer
1932 von Neumann quantum entropy / 1962 Umegaki quantumrelative entropy / 1973 Lieb–Ruskai strong subadditivity of quantumentropy / 1975 Lindblad data-processing for quantum relative entropy
1970s theory of quantum measurements and similarity measures forquantum states — Helstrom, Holevo (Shannon Award 2016), Ozawa,Bures, Uhlmann, etc.
Mark M. Wilde (LSU) 3 / 113
![Page 11: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/11.jpg)
Prehistory of quantum information theory
1927 Heisenberg uncertainty principle
1935 Einstein–Podolsky–Rosen paper questioning compatibility ofuncertainty principle and phenomenon of quantum entanglement /1964 Bell’s theorem as an answer / 2009 Berta et al. entropicuncertainty relation as another answer
1932 von Neumann quantum entropy / 1962 Umegaki quantumrelative entropy / 1973 Lieb–Ruskai strong subadditivity of quantumentropy / 1975 Lindblad data-processing for quantum relative entropy
1970s theory of quantum measurements and similarity measures forquantum states — Helstrom, Holevo (Shannon Award 2016), Ozawa,Bures, Uhlmann, etc.
Mark M. Wilde (LSU) 3 / 113
![Page 12: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/12.jpg)
Prehistory of quantum information theory
1927 Heisenberg uncertainty principle
1935 Einstein–Podolsky–Rosen paper questioning compatibility ofuncertainty principle and phenomenon of quantum entanglement /1964 Bell’s theorem as an answer / 2009 Berta et al. entropicuncertainty relation as another answer
1932 von Neumann quantum entropy / 1962 Umegaki quantumrelative entropy / 1973 Lieb–Ruskai strong subadditivity of quantumentropy / 1975 Lindblad data-processing for quantum relative entropy
1970s theory of quantum measurements and similarity measures forquantum states — Helstrom, Holevo (Shannon Award 2016), Ozawa,Bures, Uhlmann, etc.
Mark M. Wilde (LSU) 3 / 113
![Page 13: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/13.jpg)
Information theory
1948 — Shannon set the foundations of information theory, definingnotions like data compression and channel capacity and givinganswers in terms of entropy and mutual information, resp.
Shannon considered only classical physics (without quantum effects)
His work (and that of others) ultimately led to questions like:
“How do quantum effects enhance communication capacity?”
“How do quantum effects enhance communication security?”
“What are some quantum communication tasks that do not have acounterpart in the classical world?”
Mark M. Wilde (LSU) 4 / 113
![Page 14: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/14.jpg)
Information theory
1948 — Shannon set the foundations of information theory, definingnotions like data compression and channel capacity and givinganswers in terms of entropy and mutual information, resp.
Shannon considered only classical physics (without quantum effects)
His work (and that of others) ultimately led to questions like:
“How do quantum effects enhance communication capacity?”
“How do quantum effects enhance communication security?”
“What are some quantum communication tasks that do not have acounterpart in the classical world?”
Mark M. Wilde (LSU) 4 / 113
![Page 15: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/15.jpg)
Information theory
1948 — Shannon set the foundations of information theory, definingnotions like data compression and channel capacity and givinganswers in terms of entropy and mutual information, resp.
Shannon considered only classical physics (without quantum effects)
His work (and that of others) ultimately led to questions like:
“How do quantum effects enhance communication capacity?”
“How do quantum effects enhance communication security?”
“What are some quantum communication tasks that do not have acounterpart in the classical world?”
Mark M. Wilde (LSU) 4 / 113
![Page 16: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/16.jpg)
Information theory
1948 — Shannon set the foundations of information theory, definingnotions like data compression and channel capacity and givinganswers in terms of entropy and mutual information, resp.
Shannon considered only classical physics (without quantum effects)
His work (and that of others) ultimately led to questions like:
“How do quantum effects enhance communication capacity?”
“How do quantum effects enhance communication security?”
“What are some quantum communication tasks that do not have acounterpart in the classical world?”
Mark M. Wilde (LSU) 4 / 113
![Page 17: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/17.jpg)
Information theory
1948 — Shannon set the foundations of information theory, definingnotions like data compression and channel capacity and givinganswers in terms of entropy and mutual information, resp.
Shannon considered only classical physics (without quantum effects)
His work (and that of others) ultimately led to questions like:
“How do quantum effects enhance communication capacity?”
“How do quantum effects enhance communication security?”
“What are some quantum communication tasks that do not have acounterpart in the classical world?”
Mark M. Wilde (LSU) 4 / 113
![Page 18: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/18.jpg)
Information theory
1948 — Shannon set the foundations of information theory, definingnotions like data compression and channel capacity and givinganswers in terms of entropy and mutual information, resp.
Shannon considered only classical physics (without quantum effects)
His work (and that of others) ultimately led to questions like:
“How do quantum effects enhance communication capacity?”
“How do quantum effects enhance communication security?”
“What are some quantum communication tasks that do not have acounterpart in the classical world?”
Mark M. Wilde (LSU) 4 / 113
![Page 19: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/19.jpg)
Information theory
1948 — Shannon set the foundations of information theory, definingnotions like data compression and channel capacity and givinganswers in terms of entropy and mutual information, resp.
Shannon considered only classical physics (without quantum effects)
His work (and that of others) ultimately led to questions like:
“How do quantum effects enhance communication capacity?”
“How do quantum effects enhance communication security?”
“What are some quantum communication tasks that do not have acounterpart in the classical world?”
Mark M. Wilde (LSU) 4 / 113
![Page 20: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/20.jpg)
Tutorial overview
Quantum states and channels
Fundamental protocols: Bell / CHSH game, entanglementdistribution, super-dense coding, quantum teleportation
Distance measures for quantum states
Information measures
Quantum data compression
Communication over quantum channels
Mark M. Wilde (LSU) 5 / 113
![Page 21: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/21.jpg)
Tutorial overview
Quantum states and channels
Fundamental protocols: Bell / CHSH game, entanglementdistribution, super-dense coding, quantum teleportation
Distance measures for quantum states
Information measures
Quantum data compression
Communication over quantum channels
Mark M. Wilde (LSU) 5 / 113
![Page 22: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/22.jpg)
Tutorial overview
Quantum states and channels
Fundamental protocols: Bell / CHSH game, entanglementdistribution, super-dense coding, quantum teleportation
Distance measures for quantum states
Information measures
Quantum data compression
Communication over quantum channels
Mark M. Wilde (LSU) 5 / 113
![Page 23: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/23.jpg)
Tutorial overview
Quantum states and channels
Fundamental protocols: Bell / CHSH game, entanglementdistribution, super-dense coding, quantum teleportation
Distance measures for quantum states
Information measures
Quantum data compression
Communication over quantum channels
Mark M. Wilde (LSU) 5 / 113
![Page 24: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/24.jpg)
Tutorial overview
Quantum states and channels
Fundamental protocols: Bell / CHSH game, entanglementdistribution, super-dense coding, quantum teleportation
Distance measures for quantum states
Information measures
Quantum data compression
Communication over quantum channels
Mark M. Wilde (LSU) 5 / 113
![Page 25: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/25.jpg)
Tutorial overview
Quantum states and channels
Fundamental protocols: Bell / CHSH game, entanglementdistribution, super-dense coding, quantum teleportation
Distance measures for quantum states
Information measures
Quantum data compression
Communication over quantum channels
Mark M. Wilde (LSU) 5 / 113
![Page 26: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/26.jpg)
Tutorial overview
Quantum states and channels
Fundamental protocols: Bell / CHSH game, entanglementdistribution, super-dense coding, quantum teleportation
Distance measures for quantum states
Information measures
Quantum data compression
Communication over quantum channels
Mark M. Wilde (LSU) 5 / 113
![Page 27: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/27.jpg)
Review of quantum formalism
Let’s begin by reviewing some basics of quantum information
All we need to start understanding quantum information is how torepresent states and evolutions of quantum systems.
We do this by using density matrices and quantum channels.
These ideas extend how we represent states of a classical system withprobability distributions and evolutions of these classical systems withclassical channels (conditional probability distributions).
We’ll find that the set of quantum states contains all classical statesand is far richer, which is suggestive of why we can do things that arenot possible in classical information theory.
Mark M. Wilde (LSU) 6 / 113
![Page 28: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/28.jpg)
Review of quantum formalism
Let’s begin by reviewing some basics of quantum information
All we need to start understanding quantum information is how torepresent states and evolutions of quantum systems.
We do this by using density matrices and quantum channels.
These ideas extend how we represent states of a classical system withprobability distributions and evolutions of these classical systems withclassical channels (conditional probability distributions).
We’ll find that the set of quantum states contains all classical statesand is far richer, which is suggestive of why we can do things that arenot possible in classical information theory.
Mark M. Wilde (LSU) 6 / 113
![Page 29: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/29.jpg)
Review of quantum formalism
Let’s begin by reviewing some basics of quantum information
All we need to start understanding quantum information is how torepresent states and evolutions of quantum systems.
We do this by using density matrices and quantum channels.
These ideas extend how we represent states of a classical system withprobability distributions and evolutions of these classical systems withclassical channels (conditional probability distributions).
We’ll find that the set of quantum states contains all classical statesand is far richer, which is suggestive of why we can do things that arenot possible in classical information theory.
Mark M. Wilde (LSU) 6 / 113
![Page 30: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/30.jpg)
Review of quantum formalism
Let’s begin by reviewing some basics of quantum information
All we need to start understanding quantum information is how torepresent states and evolutions of quantum systems.
We do this by using density matrices and quantum channels.
These ideas extend how we represent states of a classical system withprobability distributions and evolutions of these classical systems withclassical channels (conditional probability distributions).
We’ll find that the set of quantum states contains all classical statesand is far richer, which is suggestive of why we can do things that arenot possible in classical information theory.
Mark M. Wilde (LSU) 6 / 113
![Page 31: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/31.jpg)
Review of quantum formalism
Let’s begin by reviewing some basics of quantum information
All we need to start understanding quantum information is how torepresent states and evolutions of quantum systems.
We do this by using density matrices and quantum channels.
These ideas extend how we represent states of a classical system withprobability distributions and evolutions of these classical systems withclassical channels (conditional probability distributions).
We’ll find that the set of quantum states contains all classical statesand is far richer, which is suggestive of why we can do things that arenot possible in classical information theory.
Mark M. Wilde (LSU) 6 / 113
![Page 32: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/32.jpg)
Review of quantum formalism
Let’s begin by reviewing some basics of quantum information
All we need to start understanding quantum information is how torepresent states and evolutions of quantum systems.
We do this by using density matrices and quantum channels.
These ideas extend how we represent states of a classical system withprobability distributions and evolutions of these classical systems withclassical channels (conditional probability distributions).
We’ll find that the set of quantum states contains all classical statesand is far richer, which is suggestive of why we can do things that arenot possible in classical information theory.
Mark M. Wilde (LSU) 6 / 113
![Page 33: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/33.jpg)
Quantum states
Mark M. Wilde (LSU) 7 / 113
![Page 34: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/34.jpg)
Quantum states
The state of a quantum system is given by a square matrix called thedensity matrix, usually denoted by ρ, σ, τ , ω, etc.
It should be positive semi-definite and have trace equal to one. Thatis, all of its eigenvalues should be non-negative and sum up to one.We write these conditions symbolically as ρ ≥ 0 and Tr{ρ} = 1. Canabbreviate more simply as ρ ∈ D(H), to be read as “ρ is in the set ofdensity matrices.”
The dimension of the matrix indicates the number of distinguishablestates of the quantum system.
For example, a physical qubit is a quantum system with dimensiontwo. A classical bit, which has two distinguishable states, can beembedded into a qubit.
Mark M. Wilde (LSU) 8 / 113
![Page 35: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/35.jpg)
Quantum states
The state of a quantum system is given by a square matrix called thedensity matrix, usually denoted by ρ, σ, τ , ω, etc.
It should be positive semi-definite and have trace equal to one. Thatis, all of its eigenvalues should be non-negative and sum up to one.We write these conditions symbolically as ρ ≥ 0 and Tr{ρ} = 1. Canabbreviate more simply as ρ ∈ D(H), to be read as “ρ is in the set ofdensity matrices.”
The dimension of the matrix indicates the number of distinguishablestates of the quantum system.
For example, a physical qubit is a quantum system with dimensiontwo. A classical bit, which has two distinguishable states, can beembedded into a qubit.
Mark M. Wilde (LSU) 8 / 113
![Page 36: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/36.jpg)
Quantum states
The state of a quantum system is given by a square matrix called thedensity matrix, usually denoted by ρ, σ, τ , ω, etc.
It should be positive semi-definite and have trace equal to one. Thatis, all of its eigenvalues should be non-negative and sum up to one.We write these conditions symbolically as ρ ≥ 0 and Tr{ρ} = 1. Canabbreviate more simply as ρ ∈ D(H), to be read as “ρ is in the set ofdensity matrices.”
The dimension of the matrix indicates the number of distinguishablestates of the quantum system.
For example, a physical qubit is a quantum system with dimensiontwo. A classical bit, which has two distinguishable states, can beembedded into a qubit.
Mark M. Wilde (LSU) 8 / 113
![Page 37: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/37.jpg)
Quantum states
The state of a quantum system is given by a square matrix called thedensity matrix, usually denoted by ρ, σ, τ , ω, etc.
It should be positive semi-definite and have trace equal to one. Thatis, all of its eigenvalues should be non-negative and sum up to one.We write these conditions symbolically as ρ ≥ 0 and Tr{ρ} = 1. Canabbreviate more simply as ρ ∈ D(H), to be read as “ρ is in the set ofdensity matrices.”
The dimension of the matrix indicates the number of distinguishablestates of the quantum system.
For example, a physical qubit is a quantum system with dimensiontwo. A classical bit, which has two distinguishable states, can beembedded into a qubit.
Mark M. Wilde (LSU) 8 / 113
![Page 38: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/38.jpg)
Quantum states
The state of a quantum system is given by a square matrix called thedensity matrix, usually denoted by ρ, σ, τ , ω, etc.
It should be positive semi-definite and have trace equal to one. Thatis, all of its eigenvalues should be non-negative and sum up to one.We write these conditions symbolically as ρ ≥ 0 and Tr{ρ} = 1. Canabbreviate more simply as ρ ∈ D(H), to be read as “ρ is in the set ofdensity matrices.”
The dimension of the matrix indicates the number of distinguishablestates of the quantum system.
For example, a physical qubit is a quantum system with dimensiontwo. A classical bit, which has two distinguishable states, can beembedded into a qubit.
Mark M. Wilde (LSU) 8 / 113
![Page 39: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/39.jpg)
Interpretation of density matrix
The density matrix, in addition to a description of an experimentalprocedure, is all that one requires to predict the (probabilistic)outcomes of a given experiment performed on a quantum system.
It is a generalization of (and subsumes) a probability distribution,which describes the state of a classical system. All probabilitydistributions can be embedded into a quantum state by placing theentries along the diagonal of the density matrix.
Mark M. Wilde (LSU) 9 / 113
![Page 40: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/40.jpg)
Interpretation of density matrix
The density matrix, in addition to a description of an experimentalprocedure, is all that one requires to predict the (probabilistic)outcomes of a given experiment performed on a quantum system.
It is a generalization of (and subsumes) a probability distribution,which describes the state of a classical system. All probabilitydistributions can be embedded into a quantum state by placing theentries along the diagonal of the density matrix.
Mark M. Wilde (LSU) 9 / 113
![Page 41: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/41.jpg)
Interpretation of density matrix
The density matrix, in addition to a description of an experimentalprocedure, is all that one requires to predict the (probabilistic)outcomes of a given experiment performed on a quantum system.
It is a generalization of (and subsumes) a probability distribution,which describes the state of a classical system. All probabilitydistributions can be embedded into a quantum state by placing theentries along the diagonal of the density matrix.
Mark M. Wilde (LSU) 9 / 113
![Page 42: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/42.jpg)
Let’s talk about qubits...
Superconducting phase qubit fromhttp://web.physics.ucsb.edu/˜martinisgroup/photos.shtml,
taken by Erik Lucero
Mark M. Wilde (LSU) 10 / 113
![Page 43: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/43.jpg)
Examples of quantum states
Let
|0〉 ≡[
10
], 〈0| ≡
[1 0
],
so that density matrix ρ0 ≡ |0〉〈0| =
[1 00 0
].
Similarly, let
|1〉 ≡[
01
], 〈1| ≡
[0 1
],
so that density matrix ρ1 ≡ |1〉〈1| =
[0 00 1
].
Then ρ0ρ1 = 0. The states ρ0 and ρ1 are orthogonal to each other,and, physically, this means that they are perfectly distinguishable.
What we have done here is to embed classical bits into quantum bits.We can think of ρ0 as ‘0’ and ρ1 as ‘1.’
Mark M. Wilde (LSU) 11 / 113
![Page 44: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/44.jpg)
Examples of quantum states
Let
|0〉 ≡[
10
], 〈0| ≡
[1 0
],
so that density matrix ρ0 ≡ |0〉〈0| =
[1 00 0
].
Similarly, let
|1〉 ≡[
01
], 〈1| ≡
[0 1
],
so that density matrix ρ1 ≡ |1〉〈1| =
[0 00 1
].
Then ρ0ρ1 = 0. The states ρ0 and ρ1 are orthogonal to each other,and, physically, this means that they are perfectly distinguishable.
What we have done here is to embed classical bits into quantum bits.We can think of ρ0 as ‘0’ and ρ1 as ‘1.’
Mark M. Wilde (LSU) 11 / 113
![Page 45: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/45.jpg)
Examples of quantum states
Let
|0〉 ≡[
10
], 〈0| ≡
[1 0
],
so that density matrix ρ0 ≡ |0〉〈0| =
[1 00 0
].
Similarly, let
|1〉 ≡[
01
], 〈1| ≡
[0 1
],
so that density matrix ρ1 ≡ |1〉〈1| =
[0 00 1
].
Then ρ0ρ1 = 0. The states ρ0 and ρ1 are orthogonal to each other,and, physically, this means that they are perfectly distinguishable.
What we have done here is to embed classical bits into quantum bits.We can think of ρ0 as ‘0’ and ρ1 as ‘1.’
Mark M. Wilde (LSU) 11 / 113
![Page 46: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/46.jpg)
Examples of quantum states
Let
|0〉 ≡[
10
], 〈0| ≡
[1 0
],
so that density matrix ρ0 ≡ |0〉〈0| =
[1 00 0
].
Similarly, let
|1〉 ≡[
01
], 〈1| ≡
[0 1
],
so that density matrix ρ1 ≡ |1〉〈1| =
[0 00 1
].
Then ρ0ρ1 = 0. The states ρ0 and ρ1 are orthogonal to each other,and, physically, this means that they are perfectly distinguishable.
What we have done here is to embed classical bits into quantum bits.We can think of ρ0 as ‘0’ and ρ1 as ‘1.’
Mark M. Wilde (LSU) 11 / 113
![Page 47: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/47.jpg)
Examples of quantum states
Let
|0〉 ≡[
10
], 〈0| ≡
[1 0
],
so that density matrix ρ0 ≡ |0〉〈0| =
[1 00 0
].
Similarly, let
|1〉 ≡[
01
], 〈1| ≡
[0 1
],
so that density matrix ρ1 ≡ |1〉〈1| =
[0 00 1
].
Then ρ0ρ1 = 0. The states ρ0 and ρ1 are orthogonal to each other,and, physically, this means that they are perfectly distinguishable.
What we have done here is to embed classical bits into quantum bits.We can think of ρ0 as ‘0’ and ρ1 as ‘1.’
Mark M. Wilde (LSU) 11 / 113
![Page 48: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/48.jpg)
Mixtures of quantum states
Any probabilistic mixture of two quantum states is also a quantumstate. That is, for σ0, σ1 ∈ D(H) and p ∈ [0, 1], we have
pσ0 + (1− p)σ1 ∈ D(H).
The set of density matrices is thus convex.
For our classical example, we find
pρ0 + (1− p)ρ1 = p|0〉〈0|+ (1− p)|1〉〈1|
=
[p 00 1− p
].
This is the statement that probabilistic classical bits can be embeddedinto quantum bits, and the probabilities appear along the diagonal ofthe matrix. Can we have other kinds of quantum states?
Mark M. Wilde (LSU) 12 / 113
![Page 49: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/49.jpg)
Mixtures of quantum states
Any probabilistic mixture of two quantum states is also a quantumstate. That is, for σ0, σ1 ∈ D(H) and p ∈ [0, 1], we have
pσ0 + (1− p)σ1 ∈ D(H).
The set of density matrices is thus convex.
For our classical example, we find
pρ0 + (1− p)ρ1 = p|0〉〈0|+ (1− p)|1〉〈1|
=
[p 00 1− p
].
This is the statement that probabilistic classical bits can be embeddedinto quantum bits, and the probabilities appear along the diagonal ofthe matrix. Can we have other kinds of quantum states?
Mark M. Wilde (LSU) 12 / 113
![Page 50: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/50.jpg)
Mixtures of quantum states
Any probabilistic mixture of two quantum states is also a quantumstate. That is, for σ0, σ1 ∈ D(H) and p ∈ [0, 1], we have
pσ0 + (1− p)σ1 ∈ D(H).
The set of density matrices is thus convex.
For our classical example, we find
pρ0 + (1− p)ρ1 = p|0〉〈0|+ (1− p)|1〉〈1|
=
[p 00 1− p
].
This is the statement that probabilistic classical bits can be embeddedinto quantum bits, and the probabilities appear along the diagonal ofthe matrix. Can we have other kinds of quantum states?
Mark M. Wilde (LSU) 12 / 113
![Page 51: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/51.jpg)
Mixtures of quantum states
Any probabilistic mixture of two quantum states is also a quantumstate. That is, for σ0, σ1 ∈ D(H) and p ∈ [0, 1], we have
pσ0 + (1− p)σ1 ∈ D(H).
The set of density matrices is thus convex.
For our classical example, we find
pρ0 + (1− p)ρ1 = p|0〉〈0|+ (1− p)|1〉〈1|
=
[p 00 1− p
].
This is the statement that probabilistic classical bits can be embeddedinto quantum bits, and the probabilities appear along the diagonal ofthe matrix. Can we have other kinds of quantum states?
Mark M. Wilde (LSU) 12 / 113
![Page 52: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/52.jpg)
Superpositions of quantum states
Construct the following unit vector as a superposition of |0〉 and |1〉:
|ψ〉 ≡ α|0〉+ β|1〉 =
[αβ
],
where α, β ∈ C and |α|2 + |β|2 = 1. Note that 〈ψ| =[α∗ β∗
]and
〈ϕ|ψ〉 denotes the inner product of vectors |ψ〉 and |ϕ〉.The unit vector |ψ〉 leads to the following quantum state:
|ψ〉〈ψ| =
[|α|2 αβ∗
βα∗ |β|2].
The difference between this quantum state and the others we’veconsidered so far is the presence of off-diagonal elements in thedensity matrix (called quantum coherences).
This state is physically distinct from
[|α|2 0
0 |β|2]
.
Mark M. Wilde (LSU) 13 / 113
![Page 53: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/53.jpg)
Superpositions of quantum states
Construct the following unit vector as a superposition of |0〉 and |1〉:
|ψ〉 ≡ α|0〉+ β|1〉 =
[αβ
],
where α, β ∈ C and |α|2 + |β|2 = 1. Note that 〈ψ| =[α∗ β∗
]and
〈ϕ|ψ〉 denotes the inner product of vectors |ψ〉 and |ϕ〉.
The unit vector |ψ〉 leads to the following quantum state:
|ψ〉〈ψ| =
[|α|2 αβ∗
βα∗ |β|2].
The difference between this quantum state and the others we’veconsidered so far is the presence of off-diagonal elements in thedensity matrix (called quantum coherences).
This state is physically distinct from
[|α|2 0
0 |β|2]
.
Mark M. Wilde (LSU) 13 / 113
![Page 54: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/54.jpg)
Superpositions of quantum states
Construct the following unit vector as a superposition of |0〉 and |1〉:
|ψ〉 ≡ α|0〉+ β|1〉 =
[αβ
],
where α, β ∈ C and |α|2 + |β|2 = 1. Note that 〈ψ| =[α∗ β∗
]and
〈ϕ|ψ〉 denotes the inner product of vectors |ψ〉 and |ϕ〉.The unit vector |ψ〉 leads to the following quantum state:
|ψ〉〈ψ| =
[|α|2 αβ∗
βα∗ |β|2].
The difference between this quantum state and the others we’veconsidered so far is the presence of off-diagonal elements in thedensity matrix (called quantum coherences).
This state is physically distinct from
[|α|2 0
0 |β|2]
.
Mark M. Wilde (LSU) 13 / 113
![Page 55: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/55.jpg)
Superpositions of quantum states
Construct the following unit vector as a superposition of |0〉 and |1〉:
|ψ〉 ≡ α|0〉+ β|1〉 =
[αβ
],
where α, β ∈ C and |α|2 + |β|2 = 1. Note that 〈ψ| =[α∗ β∗
]and
〈ϕ|ψ〉 denotes the inner product of vectors |ψ〉 and |ϕ〉.The unit vector |ψ〉 leads to the following quantum state:
|ψ〉〈ψ| =
[|α|2 αβ∗
βα∗ |β|2].
The difference between this quantum state and the others we’veconsidered so far is the presence of off-diagonal elements in thedensity matrix (called quantum coherences).
This state is physically distinct from
[|α|2 0
0 |β|2]
.
Mark M. Wilde (LSU) 13 / 113
![Page 56: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/56.jpg)
Superpositions of quantum states
Construct the following unit vector as a superposition of |0〉 and |1〉:
|ψ〉 ≡ α|0〉+ β|1〉 =
[αβ
],
where α, β ∈ C and |α|2 + |β|2 = 1. Note that 〈ψ| =[α∗ β∗
]and
〈ϕ|ψ〉 denotes the inner product of vectors |ψ〉 and |ϕ〉.The unit vector |ψ〉 leads to the following quantum state:
|ψ〉〈ψ| =
[|α|2 αβ∗
βα∗ |β|2].
The difference between this quantum state and the others we’veconsidered so far is the presence of off-diagonal elements in thedensity matrix (called quantum coherences).
This state is physically distinct from
[|α|2 0
0 |β|2]
.
Mark M. Wilde (LSU) 13 / 113
![Page 57: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/57.jpg)
Bloch sphere
We can visualize the state of a qubit using the Bloch sphere. To seethis, consider the Pauli matrices
I ≡[
1 00 1
], X ≡
[0 11 0
], Y ≡
[0 −ii 0
], Z ≡
[1 00 −1
].
The last three Pauli matrices have eigenvalues ±1 and eigenvectors:
|±〉 ≡ 1√2
(|0〉 ± |1〉) , |±Y 〉 ≡1√2
(|0〉 ± i |1〉) , |0〉, |1〉.
We can write the density matrix ρ of a qubit in terms of threeparameters rx , ry , and rz :
ρ =1
2(I + rxX + ryY + rzZ ) ,
where r2x + r2
y + r2z ≤ 1, which is the equation of a unit sphere in R3.
Mark M. Wilde (LSU) 14 / 113
![Page 58: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/58.jpg)
Bloch sphere
We can visualize the state of a qubit using the Bloch sphere. To seethis, consider the Pauli matrices
I ≡[
1 00 1
], X ≡
[0 11 0
], Y ≡
[0 −ii 0
], Z ≡
[1 00 −1
].
The last three Pauli matrices have eigenvalues ±1 and eigenvectors:
|±〉 ≡ 1√2
(|0〉 ± |1〉) , |±Y 〉 ≡1√2
(|0〉 ± i |1〉) , |0〉, |1〉.
We can write the density matrix ρ of a qubit in terms of threeparameters rx , ry , and rz :
ρ =1
2(I + rxX + ryY + rzZ ) ,
where r2x + r2
y + r2z ≤ 1, which is the equation of a unit sphere in R3.
Mark M. Wilde (LSU) 14 / 113
![Page 59: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/59.jpg)
Bloch sphere
We can visualize the state of a qubit using the Bloch sphere. To seethis, consider the Pauli matrices
I ≡[
1 00 1
], X ≡
[0 11 0
], Y ≡
[0 −ii 0
], Z ≡
[1 00 −1
].
The last three Pauli matrices have eigenvalues ±1 and eigenvectors:
|±〉 ≡ 1√2
(|0〉 ± |1〉) , |±Y 〉 ≡1√2
(|0〉 ± i |1〉) , |0〉, |1〉.
We can write the density matrix ρ of a qubit in terms of threeparameters rx , ry , and rz :
ρ =1
2(I + rxX + ryY + rzZ ) ,
where r2x + r2
y + r2z ≤ 1, which is the equation of a unit sphere in R3.
Mark M. Wilde (LSU) 14 / 113
![Page 60: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/60.jpg)
Bloch sphere
We can visualize the state of a qubit using the Bloch sphere:
The maximally mixed state I/2 = (|0〉〈0|+ |1〉〈1|)/2 is at the center.
Classical states are on the line going from |0〉 to |1〉.A quantum state is pure if it is on the surface and otherwise mixed.
Mark M. Wilde (LSU) 15 / 113
![Page 61: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/61.jpg)
Bloch sphere
We can visualize the state of a qubit using the Bloch sphere:
The maximally mixed state I/2 = (|0〉〈0|+ |1〉〈1|)/2 is at the center.
Classical states are on the line going from |0〉 to |1〉.A quantum state is pure if it is on the surface and otherwise mixed.
Mark M. Wilde (LSU) 15 / 113
![Page 62: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/62.jpg)
Bloch sphere
We can visualize the state of a qubit using the Bloch sphere:
The maximally mixed state I/2 = (|0〉〈0|+ |1〉〈1|)/2 is at the center.
Classical states are on the line going from |0〉 to |1〉.
A quantum state is pure if it is on the surface and otherwise mixed.
Mark M. Wilde (LSU) 15 / 113
![Page 63: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/63.jpg)
Bloch sphere
We can visualize the state of a qubit using the Bloch sphere:
The maximally mixed state I/2 = (|0〉〈0|+ |1〉〈1|)/2 is at the center.
Classical states are on the line going from |0〉 to |1〉.A quantum state is pure if it is on the surface and otherwise mixed.
Mark M. Wilde (LSU) 15 / 113
![Page 64: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/64.jpg)
Higher dimensional quantum systems
A density matrix can have dimension ≥ 2 and can be written as
ρ =∑i ,j
ρi ,j |i〉〈j |,
where {|i〉 ≡ ei} is the standard basis and ρi ,j are the matrix elements.
Since every density matrix is positive semi-definite and has traceequal to one, it has a spectral decomposition as
ρ =∑x
pX (x)|φx〉〈φx |,
where {pX (x)} are the non-negative eigenvalues, summing to one,and {|φx〉} is a set of orthonormal eigenvectors.
A density matrix ρ is pure if there exists a unit vector |ψ〉 such thatρ = |ψ〉〈ψ| and otherwise it is mixed.
Mark M. Wilde (LSU) 16 / 113
![Page 65: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/65.jpg)
Higher dimensional quantum systems
A density matrix can have dimension ≥ 2 and can be written as
ρ =∑i ,j
ρi ,j |i〉〈j |,
where {|i〉 ≡ ei} is the standard basis and ρi ,j are the matrix elements.
Since every density matrix is positive semi-definite and has traceequal to one, it has a spectral decomposition as
ρ =∑x
pX (x)|φx〉〈φx |,
where {pX (x)} are the non-negative eigenvalues, summing to one,and {|φx〉} is a set of orthonormal eigenvectors.
A density matrix ρ is pure if there exists a unit vector |ψ〉 such thatρ = |ψ〉〈ψ| and otherwise it is mixed.
Mark M. Wilde (LSU) 16 / 113
![Page 66: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/66.jpg)
Higher dimensional quantum systems
A density matrix can have dimension ≥ 2 and can be written as
ρ =∑i ,j
ρi ,j |i〉〈j |,
where {|i〉 ≡ ei} is the standard basis and ρi ,j are the matrix elements.
Since every density matrix is positive semi-definite and has traceequal to one, it has a spectral decomposition as
ρ =∑x
pX (x)|φx〉〈φx |,
where {pX (x)} are the non-negative eigenvalues, summing to one,and {|φx〉} is a set of orthonormal eigenvectors.
A density matrix ρ is pure if there exists a unit vector |ψ〉 such thatρ = |ψ〉〈ψ| and otherwise it is mixed.
Mark M. Wilde (LSU) 16 / 113
![Page 67: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/67.jpg)
Higher dimensional quantum systems
A density matrix can have dimension ≥ 2 and can be written as
ρ =∑i ,j
ρi ,j |i〉〈j |,
where {|i〉 ≡ ei} is the standard basis and ρi ,j are the matrix elements.
Since every density matrix is positive semi-definite and has traceequal to one, it has a spectral decomposition as
ρ =∑x
pX (x)|φx〉〈φx |,
where {pX (x)} are the non-negative eigenvalues, summing to one,and {|φx〉} is a set of orthonormal eigenvectors.
A density matrix ρ is pure if there exists a unit vector |ψ〉 such thatρ = |ψ〉〈ψ| and otherwise it is mixed.
Mark M. Wilde (LSU) 16 / 113
![Page 68: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/68.jpg)
Multiple qubits...
IBM five-qubit universal quantum computer (released May 2016)
Mark M. Wilde (LSU) 17 / 113
![Page 69: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/69.jpg)
Composite quantum systems
Just as we need more than one bit for information processing tobecome interesting, quantum information really only becomesinteresting when multiple quantum systems can interact.
We use Cartesian product to represent state of two or more bits:
(0, 0), (0, 1), (1, 0), (1, 1) ∈ Z2 × Z2,
but Cartesian product is not rich enough to capture quantum states.
Consider that before we constructed a quantum state from asuperposition of two unit vectors. So we could imagine constructing aquantum state from a superposition of vectors as
α|0, 0〉+ β|0, 1〉+ γ|1, 0〉+ δ|1, 1〉,
where |α|2 + |β|2 + |γ|2 + |δ|2 = 1. But what are |i , j〉?
Mark M. Wilde (LSU) 18 / 113
![Page 70: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/70.jpg)
Composite quantum systems
Just as we need more than one bit for information processing tobecome interesting, quantum information really only becomesinteresting when multiple quantum systems can interact.
We use Cartesian product to represent state of two or more bits:
(0, 0), (0, 1), (1, 0), (1, 1) ∈ Z2 × Z2,
but Cartesian product is not rich enough to capture quantum states.
Consider that before we constructed a quantum state from asuperposition of two unit vectors. So we could imagine constructing aquantum state from a superposition of vectors as
α|0, 0〉+ β|0, 1〉+ γ|1, 0〉+ δ|1, 1〉,
where |α|2 + |β|2 + |γ|2 + |δ|2 = 1. But what are |i , j〉?
Mark M. Wilde (LSU) 18 / 113
![Page 71: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/71.jpg)
Composite quantum systems
Just as we need more than one bit for information processing tobecome interesting, quantum information really only becomesinteresting when multiple quantum systems can interact.
We use Cartesian product to represent state of two or more bits:
(0, 0), (0, 1), (1, 0), (1, 1) ∈ Z2 × Z2,
but Cartesian product is not rich enough to capture quantum states.
Consider that before we constructed a quantum state from asuperposition of two unit vectors. So we could imagine constructing aquantum state from a superposition of vectors as
α|0, 0〉+ β|0, 1〉+ γ|1, 0〉+ δ|1, 1〉,
where |α|2 + |β|2 + |γ|2 + |δ|2 = 1. But what are |i , j〉?
Mark M. Wilde (LSU) 18 / 113
![Page 72: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/72.jpg)
Composite quantum systems
Just as we need more than one bit for information processing tobecome interesting, quantum information really only becomesinteresting when multiple quantum systems can interact.
We use Cartesian product to represent state of two or more bits:
(0, 0), (0, 1), (1, 0), (1, 1) ∈ Z2 × Z2,
but Cartesian product is not rich enough to capture quantum states.
Consider that before we constructed a quantum state from asuperposition of two unit vectors. So we could imagine constructing aquantum state from a superposition of vectors as
α|0, 0〉+ β|0, 1〉+ γ|1, 0〉+ δ|1, 1〉,
where |α|2 + |β|2 + |γ|2 + |δ|2 = 1. But what are |i , j〉?
Mark M. Wilde (LSU) 18 / 113
![Page 73: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/73.jpg)
Tensor product
We use the tensor product to represent multiple quantum systems.
For vectors, it is defined as
[a1
b1
]⊗[
a2
b2
]≡
a1
[a2
b2
]b1
[a2
b2
] =
a1a2
a1b2
b1a2
b1b2
.So, then with this definition, we have
|ϕ〉 ≡ α|0〉 ⊗ |0〉+ β|0〉 ⊗ |1〉+ γ|1〉 ⊗ |0〉+ δ|1〉 ⊗ |1〉 =
αβγδ
,which leads to a two-qubit density operator |ϕ〉〈ϕ|.
Mark M. Wilde (LSU) 19 / 113
![Page 74: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/74.jpg)
Tensor product
We use the tensor product to represent multiple quantum systems.
For vectors, it is defined as
[a1
b1
]⊗[
a2
b2
]≡
a1
[a2
b2
]b1
[a2
b2
] =
a1a2
a1b2
b1a2
b1b2
.So, then with this definition, we have
|ϕ〉 ≡ α|0〉 ⊗ |0〉+ β|0〉 ⊗ |1〉+ γ|1〉 ⊗ |0〉+ δ|1〉 ⊗ |1〉 =
αβγδ
,which leads to a two-qubit density operator |ϕ〉〈ϕ|.
Mark M. Wilde (LSU) 19 / 113
![Page 75: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/75.jpg)
Tensor product
We use the tensor product to represent multiple quantum systems.
For vectors, it is defined as
[a1
b1
]⊗[
a2
b2
]≡
a1
[a2
b2
]b1
[a2
b2
] =
a1a2
a1b2
b1a2
b1b2
.
So, then with this definition, we have
|ϕ〉 ≡ α|0〉 ⊗ |0〉+ β|0〉 ⊗ |1〉+ γ|1〉 ⊗ |0〉+ δ|1〉 ⊗ |1〉 =
αβγδ
,which leads to a two-qubit density operator |ϕ〉〈ϕ|.
Mark M. Wilde (LSU) 19 / 113
![Page 76: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/76.jpg)
Tensor product
We use the tensor product to represent multiple quantum systems.
For vectors, it is defined as
[a1
b1
]⊗[
a2
b2
]≡
a1
[a2
b2
]b1
[a2
b2
] =
a1a2
a1b2
b1a2
b1b2
.So, then with this definition, we have
|ϕ〉 ≡ α|0〉 ⊗ |0〉+ β|0〉 ⊗ |1〉+ γ|1〉 ⊗ |0〉+ δ|1〉 ⊗ |1〉 =
αβγδ
,which leads to a two-qubit density operator |ϕ〉〈ϕ|.
Mark M. Wilde (LSU) 19 / 113
![Page 77: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/77.jpg)
System labels
Often it can be helpful to write system labels, which indicate whichqubit Alice possesses and which Bob possesses:
|ϕ〉AB ≡ α|0〉A ⊗ |0〉B + β|0〉A ⊗ |1〉B + γ|1〉A ⊗ |0〉B + δ|1〉A ⊗ |1〉B .
We can also write the labels on the two-qubit density operator:
|ϕ〉〈ϕ|AB .
Often we abbreviate the above more simply as
α|00〉AB + β|01〉AB + γ|10〉AB + δ|11〉AB .
Mark M. Wilde (LSU) 20 / 113
![Page 78: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/78.jpg)
System labels
Often it can be helpful to write system labels, which indicate whichqubit Alice possesses and which Bob possesses:
|ϕ〉AB ≡ α|0〉A ⊗ |0〉B + β|0〉A ⊗ |1〉B + γ|1〉A ⊗ |0〉B + δ|1〉A ⊗ |1〉B .
We can also write the labels on the two-qubit density operator:
|ϕ〉〈ϕ|AB .
Often we abbreviate the above more simply as
α|00〉AB + β|01〉AB + γ|10〉AB + δ|11〉AB .
Mark M. Wilde (LSU) 20 / 113
![Page 79: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/79.jpg)
System labels
Often it can be helpful to write system labels, which indicate whichqubit Alice possesses and which Bob possesses:
|ϕ〉AB ≡ α|0〉A ⊗ |0〉B + β|0〉A ⊗ |1〉B + γ|1〉A ⊗ |0〉B + δ|1〉A ⊗ |1〉B .
We can also write the labels on the two-qubit density operator:
|ϕ〉〈ϕ|AB .
Often we abbreviate the above more simply as
α|00〉AB + β|01〉AB + γ|10〉AB + δ|11〉AB .
Mark M. Wilde (LSU) 20 / 113
![Page 80: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/80.jpg)
Tensor product for matrices
For matrices K and L, the tensor product is defined in a similar way:
K ⊗ L ≡[k11 k12
k21 k22
]⊗[l11 l12
l21 l22
]
≡
k11
[l11 l12
l21 l22
]k12
[l11 l12
l21 l22
]k21
[l11 l12
l21 l22
]k22
[l11 l12
l21 l22
]
=
k11l11 k11l12 k12l11 k12l12
k11l21 k11l22 k12l21 k12l22
k21l11 k21l12 k22l11 k22l12
k21l21 k21l22 k22l21 k22l22
.
Mark M. Wilde (LSU) 21 / 113
![Page 81: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/81.jpg)
Tensor product for matrices
For matrices K and L, the tensor product is defined in a similar way:
K ⊗ L ≡[k11 k12
k21 k22
]⊗[l11 l12
l21 l22
]
≡
k11
[l11 l12
l21 l22
]k12
[l11 l12
l21 l22
]k21
[l11 l12
l21 l22
]k22
[l11 l12
l21 l22
]
=
k11l11 k11l12 k12l11 k12l12
k11l21 k11l22 k12l21 k12l22
k21l11 k21l12 k22l11 k22l12
k21l21 k21l22 k22l21 k22l22
.
Mark M. Wilde (LSU) 21 / 113
![Page 82: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/82.jpg)
Properties of tensor product
For vectors:
z(|φ〉 ⊗ |ψ〉) = (z |φ〉)⊗ |ψ〉 = |φ〉 ⊗ (z |ψ〉),(|φ1〉+ |φ2〉)⊗ |ψ〉 = |φ1〉 ⊗ |ψ〉+ |φ2〉 ⊗ |ψ〉,|φ〉 ⊗ (|ψ1〉+ |ψ2〉) = |φ〉 ⊗ |ψ1〉+ |φ〉 ⊗ |ψ2〉.
Matrices acting on vectors:
(K ⊗ L)(|φ〉 ⊗ |ψ〉) = K |φ〉 ⊗ L|ψ〉,
(K ⊗ L)
(∑x
λx |φx〉 ⊗ |ψx〉
)=∑x
λxK |φx〉 ⊗ L|ψx〉,(∑x
µxKx ⊗ Lx
)(|φ〉 ⊗ |ψ〉) =
∑x
µxKx |φ〉 ⊗ Lx |ψ〉.
Inner product: (〈φ1| ⊗ 〈ψ1|)(|φ2〉 ⊗ |ψ2〉) = 〈φ1|φ2〉〈ψ1|ψ2〉.
Mark M. Wilde (LSU) 22 / 113
![Page 83: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/83.jpg)
Properties of tensor product
For vectors:
z(|φ〉 ⊗ |ψ〉) = (z |φ〉)⊗ |ψ〉 = |φ〉 ⊗ (z |ψ〉),(|φ1〉+ |φ2〉)⊗ |ψ〉 = |φ1〉 ⊗ |ψ〉+ |φ2〉 ⊗ |ψ〉,|φ〉 ⊗ (|ψ1〉+ |ψ2〉) = |φ〉 ⊗ |ψ1〉+ |φ〉 ⊗ |ψ2〉.
Matrices acting on vectors:
(K ⊗ L)(|φ〉 ⊗ |ψ〉) = K |φ〉 ⊗ L|ψ〉,
(K ⊗ L)
(∑x
λx |φx〉 ⊗ |ψx〉
)=∑x
λxK |φx〉 ⊗ L|ψx〉,(∑x
µxKx ⊗ Lx
)(|φ〉 ⊗ |ψ〉) =
∑x
µxKx |φ〉 ⊗ Lx |ψ〉.
Inner product: (〈φ1| ⊗ 〈ψ1|)(|φ2〉 ⊗ |ψ2〉) = 〈φ1|φ2〉〈ψ1|ψ2〉.
Mark M. Wilde (LSU) 22 / 113
![Page 84: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/84.jpg)
Properties of tensor product
For vectors:
z(|φ〉 ⊗ |ψ〉) = (z |φ〉)⊗ |ψ〉 = |φ〉 ⊗ (z |ψ〉),(|φ1〉+ |φ2〉)⊗ |ψ〉 = |φ1〉 ⊗ |ψ〉+ |φ2〉 ⊗ |ψ〉,|φ〉 ⊗ (|ψ1〉+ |ψ2〉) = |φ〉 ⊗ |ψ1〉+ |φ〉 ⊗ |ψ2〉.
Matrices acting on vectors:
(K ⊗ L)(|φ〉 ⊗ |ψ〉) = K |φ〉 ⊗ L|ψ〉,
(K ⊗ L)
(∑x
λx |φx〉 ⊗ |ψx〉
)=∑x
λxK |φx〉 ⊗ L|ψx〉,(∑x
µxKx ⊗ Lx
)(|φ〉 ⊗ |ψ〉) =
∑x
µxKx |φ〉 ⊗ Lx |ψ〉.
Inner product: (〈φ1| ⊗ 〈ψ1|)(|φ2〉 ⊗ |ψ2〉) = 〈φ1|φ2〉〈ψ1|ψ2〉.
Mark M. Wilde (LSU) 22 / 113
![Page 85: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/85.jpg)
Properties of tensor product
For vectors:
z(|φ〉 ⊗ |ψ〉) = (z |φ〉)⊗ |ψ〉 = |φ〉 ⊗ (z |ψ〉),(|φ1〉+ |φ2〉)⊗ |ψ〉 = |φ1〉 ⊗ |ψ〉+ |φ2〉 ⊗ |ψ〉,|φ〉 ⊗ (|ψ1〉+ |ψ2〉) = |φ〉 ⊗ |ψ1〉+ |φ〉 ⊗ |ψ2〉.
Matrices acting on vectors:
(K ⊗ L)(|φ〉 ⊗ |ψ〉) = K |φ〉 ⊗ L|ψ〉,
(K ⊗ L)
(∑x
λx |φx〉 ⊗ |ψx〉
)=∑x
λxK |φx〉 ⊗ L|ψx〉,(∑x
µxKx ⊗ Lx
)(|φ〉 ⊗ |ψ〉) =
∑x
µxKx |φ〉 ⊗ Lx |ψ〉.
Inner product: (〈φ1| ⊗ 〈ψ1|)(|φ2〉 ⊗ |ψ2〉) = 〈φ1|φ2〉〈ψ1|ψ2〉.Mark M. Wilde (LSU) 22 / 113
![Page 86: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/86.jpg)
Composite quantum systems
If the state of Alice’s system is ρ and the state of Bob’s system is σand they have never interacted in the past, then the state of the jointAlice-Bob system is
ρA ⊗ σB .
We use the system labels to say who has what.
For example, their state could be
|0〉〈0|A ⊗ |0〉〈0|B , or
|1〉〈1|A ⊗ |1〉〈1|B ,
or a mixture of both, with p ∈ [0, 1]:
p|0〉〈0|A ⊗ |0〉〈0|B + (1− p)|1〉〈1|A ⊗ |1〉〈1|B .
Mark M. Wilde (LSU) 23 / 113
![Page 87: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/87.jpg)
Composite quantum systems
If the state of Alice’s system is ρ and the state of Bob’s system is σand they have never interacted in the past, then the state of the jointAlice-Bob system is
ρA ⊗ σB .
We use the system labels to say who has what.
For example, their state could be
|0〉〈0|A ⊗ |0〉〈0|B , or
|1〉〈1|A ⊗ |1〉〈1|B ,
or a mixture of both, with p ∈ [0, 1]:
p|0〉〈0|A ⊗ |0〉〈0|B + (1− p)|1〉〈1|A ⊗ |1〉〈1|B .
Mark M. Wilde (LSU) 23 / 113
![Page 88: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/88.jpg)
Composite quantum systems
If the state of Alice’s system is ρ and the state of Bob’s system is σand they have never interacted in the past, then the state of the jointAlice-Bob system is
ρA ⊗ σB .
We use the system labels to say who has what.
For example, their state could be
|0〉〈0|A ⊗ |0〉〈0|B , or
|1〉〈1|A ⊗ |1〉〈1|B ,
or a mixture of both, with p ∈ [0, 1]:
p|0〉〈0|A ⊗ |0〉〈0|B + (1− p)|1〉〈1|A ⊗ |1〉〈1|B .
Mark M. Wilde (LSU) 23 / 113
![Page 89: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/89.jpg)
Composite quantum systems
If the state of Alice’s system is ρ and the state of Bob’s system is σand they have never interacted in the past, then the state of the jointAlice-Bob system is
ρA ⊗ σB .
We use the system labels to say who has what.
For example, their state could be
|0〉〈0|A ⊗ |0〉〈0|B , or
|1〉〈1|A ⊗ |1〉〈1|B ,
or a mixture of both, with p ∈ [0, 1]:
p|0〉〈0|A ⊗ |0〉〈0|B + (1− p)|1〉〈1|A ⊗ |1〉〈1|B .
Mark M. Wilde (LSU) 23 / 113
![Page 90: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/90.jpg)
Quantum entanglement...
Depiction of quantum entanglement taken fromhttp://thelifeofpsi.com/2013/10/28/bertlmanns-socks/
Mark M. Wilde (LSU) 24 / 113
![Page 91: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/91.jpg)
Separable states and entangled states
If Alice and Bob prepare states ρxA and σxB based on a randomvariable X with distribution pX , then the state of their systems is∑
x
pX (x)ρxA ⊗ σxB .
Such states are called separable states and can be prepared usinglocal operations and classical communication (no need for a quantuminteraction between A and B to prepare these states).
By spectral decomposition, every separable state can be written as∑z
pZ (z)|ψz〉〈ψz |A ⊗ |φz〉〈φz |B ,
where, for each z , |ψz〉A and |φz〉B are unit vectors.
Entangled states are states that cannot be written in the above form.
Mark M. Wilde (LSU) 25 / 113
![Page 92: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/92.jpg)
Separable states and entangled states
If Alice and Bob prepare states ρxA and σxB based on a randomvariable X with distribution pX , then the state of their systems is∑
x
pX (x)ρxA ⊗ σxB .
Such states are called separable states and can be prepared usinglocal operations and classical communication (no need for a quantuminteraction between A and B to prepare these states).
By spectral decomposition, every separable state can be written as∑z
pZ (z)|ψz〉〈ψz |A ⊗ |φz〉〈φz |B ,
where, for each z , |ψz〉A and |φz〉B are unit vectors.
Entangled states are states that cannot be written in the above form.
Mark M. Wilde (LSU) 25 / 113
![Page 93: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/93.jpg)
Separable states and entangled states
If Alice and Bob prepare states ρxA and σxB based on a randomvariable X with distribution pX , then the state of their systems is∑
x
pX (x)ρxA ⊗ σxB .
Such states are called separable states and can be prepared usinglocal operations and classical communication (no need for a quantuminteraction between A and B to prepare these states).
By spectral decomposition, every separable state can be written as∑z
pZ (z)|ψz〉〈ψz |A ⊗ |φz〉〈φz |B ,
where, for each z , |ψz〉A and |φz〉B are unit vectors.
Entangled states are states that cannot be written in the above form.
Mark M. Wilde (LSU) 25 / 113
![Page 94: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/94.jpg)
Separable states and entangled states
If Alice and Bob prepare states ρxA and σxB based on a randomvariable X with distribution pX , then the state of their systems is∑
x
pX (x)ρxA ⊗ σxB .
Such states are called separable states and can be prepared usinglocal operations and classical communication (no need for a quantuminteraction between A and B to prepare these states).
By spectral decomposition, every separable state can be written as∑z
pZ (z)|ψz〉〈ψz |A ⊗ |φz〉〈φz |B ,
where, for each z , |ψz〉A and |φz〉B are unit vectors.
Entangled states are states that cannot be written in the above form.
Mark M. Wilde (LSU) 25 / 113
![Page 95: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/95.jpg)
Separable states and entangled states
If Alice and Bob prepare states ρxA and σxB based on a randomvariable X with distribution pX , then the state of their systems is∑
x
pX (x)ρxA ⊗ σxB .
Such states are called separable states and can be prepared usinglocal operations and classical communication (no need for a quantuminteraction between A and B to prepare these states).
By spectral decomposition, every separable state can be written as∑z
pZ (z)|ψz〉〈ψz |A ⊗ |φz〉〈φz |B ,
where, for each z , |ψz〉A and |φz〉B are unit vectors.
Entangled states are states that cannot be written in the above form.
Mark M. Wilde (LSU) 25 / 113
![Page 96: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/96.jpg)
Example of entangled state
A prominent example of an entangled state is the ebit (eee · bit):
|Φ〉〈Φ|AB ,
where |Φ〉AB ≡ 1√2
(|00〉AB + |11〉AB).
In matrix form, this is
|Φ〉〈Φ|AB =1
2
1 0 0 10 0 0 00 0 0 01 0 0 1
.To see that this is entangled, consider that for every |ψ〉A and |φ〉B
|〈Φ|AB |ψ〉A ⊗ |φ〉B |2 ≤1
2
⇒ impossible to write |Φ〉〈Φ|AB as a separable state.
Mark M. Wilde (LSU) 26 / 113
![Page 97: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/97.jpg)
Example of entangled state
A prominent example of an entangled state is the ebit (eee · bit):
|Φ〉〈Φ|AB ,
where |Φ〉AB ≡ 1√2
(|00〉AB + |11〉AB).
In matrix form, this is
|Φ〉〈Φ|AB =1
2
1 0 0 10 0 0 00 0 0 01 0 0 1
.To see that this is entangled, consider that for every |ψ〉A and |φ〉B
|〈Φ|AB |ψ〉A ⊗ |φ〉B |2 ≤1
2
⇒ impossible to write |Φ〉〈Φ|AB as a separable state.
Mark M. Wilde (LSU) 26 / 113
![Page 98: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/98.jpg)
Example of entangled state
A prominent example of an entangled state is the ebit (eee · bit):
|Φ〉〈Φ|AB ,
where |Φ〉AB ≡ 1√2
(|00〉AB + |11〉AB).
In matrix form, this is
|Φ〉〈Φ|AB =1
2
1 0 0 10 0 0 00 0 0 01 0 0 1
.
To see that this is entangled, consider that for every |ψ〉A and |φ〉B
|〈Φ|AB |ψ〉A ⊗ |φ〉B |2 ≤1
2
⇒ impossible to write |Φ〉〈Φ|AB as a separable state.
Mark M. Wilde (LSU) 26 / 113
![Page 99: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/99.jpg)
Example of entangled state
A prominent example of an entangled state is the ebit (eee · bit):
|Φ〉〈Φ|AB ,
where |Φ〉AB ≡ 1√2
(|00〉AB + |11〉AB).
In matrix form, this is
|Φ〉〈Φ|AB =1
2
1 0 0 10 0 0 00 0 0 01 0 0 1
.To see that this is entangled, consider that for every |ψ〉A and |φ〉B
|〈Φ|AB |ψ〉A ⊗ |φ〉B |2 ≤1
2
⇒ impossible to write |Φ〉〈Φ|AB as a separable state.
Mark M. Wilde (LSU) 26 / 113
![Page 100: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/100.jpg)
Example of entangled state
A prominent example of an entangled state is the ebit (eee · bit):
|Φ〉〈Φ|AB ,
where |Φ〉AB ≡ 1√2
(|00〉AB + |11〉AB).
In matrix form, this is
|Φ〉〈Φ|AB =1
2
1 0 0 10 0 0 00 0 0 01 0 0 1
.To see that this is entangled, consider that for every |ψ〉A and |φ〉B
|〈Φ|AB |ψ〉A ⊗ |φ〉B |2 ≤1
2
⇒ impossible to write |Φ〉〈Φ|AB as a separable state.
Mark M. Wilde (LSU) 26 / 113
![Page 101: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/101.jpg)
Tool: Schmidt decomposition
Schmidt decomposition theorem
Given a two-party unit vector |ψ〉AB ∈ HA ⊗HB , we can express it as
|ψ〉AB ≡d−1∑i=0
√pi |i〉A |i〉B , where
probabilities pi are real, strictly positive, and normalized∑
i pi = 1.
{|i〉A} and {|i〉B} are orthonormal bases for systems A and B.[√pi]i∈{0,...,d−1} is the vector of Schmidt coefficients.
Schmidt rank d of |ψ〉AB is equal to the number of Schmidtcoefficients pi in its Schmidt decomposition and satisfies
d ≤ min {dim(HA), dim(HB)} .
State |ψ〉〈ψ|AB is entangled iff d ≥ 2.
Mark M. Wilde (LSU) 27 / 113
![Page 102: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/102.jpg)
Tool: Schmidt decomposition
Schmidt decomposition theorem
Given a two-party unit vector |ψ〉AB ∈ HA ⊗HB , we can express it as
|ψ〉AB ≡d−1∑i=0
√pi |i〉A |i〉B , where
probabilities pi are real, strictly positive, and normalized∑
i pi = 1.
{|i〉A} and {|i〉B} are orthonormal bases for systems A and B.[√pi]i∈{0,...,d−1} is the vector of Schmidt coefficients.
Schmidt rank d of |ψ〉AB is equal to the number of Schmidtcoefficients pi in its Schmidt decomposition and satisfies
d ≤ min {dim(HA), dim(HB)} .
State |ψ〉〈ψ|AB is entangled iff d ≥ 2.
Mark M. Wilde (LSU) 27 / 113
![Page 103: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/103.jpg)
Tool: Schmidt decomposition
Schmidt decomposition theorem
Given a two-party unit vector |ψ〉AB ∈ HA ⊗HB , we can express it as
|ψ〉AB ≡d−1∑i=0
√pi |i〉A |i〉B , where
probabilities pi are real, strictly positive, and normalized∑
i pi = 1.
{|i〉A} and {|i〉B} are orthonormal bases for systems A and B.[√pi]i∈{0,...,d−1} is the vector of Schmidt coefficients.
Schmidt rank d of |ψ〉AB is equal to the number of Schmidtcoefficients pi in its Schmidt decomposition and satisfies
d ≤ min {dim(HA), dim(HB)} .
State |ψ〉〈ψ|AB is entangled iff d ≥ 2.
Mark M. Wilde (LSU) 27 / 113
![Page 104: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/104.jpg)
Tool: Schmidt decomposition
Schmidt decomposition theorem
Given a two-party unit vector |ψ〉AB ∈ HA ⊗HB , we can express it as
|ψ〉AB ≡d−1∑i=0
√pi |i〉A |i〉B , where
probabilities pi are real, strictly positive, and normalized∑
i pi = 1.
{|i〉A} and {|i〉B} are orthonormal bases for systems A and B.
[√pi]i∈{0,...,d−1} is the vector of Schmidt coefficients.
Schmidt rank d of |ψ〉AB is equal to the number of Schmidtcoefficients pi in its Schmidt decomposition and satisfies
d ≤ min {dim(HA), dim(HB)} .
State |ψ〉〈ψ|AB is entangled iff d ≥ 2.
Mark M. Wilde (LSU) 27 / 113
![Page 105: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/105.jpg)
Tool: Schmidt decomposition
Schmidt decomposition theorem
Given a two-party unit vector |ψ〉AB ∈ HA ⊗HB , we can express it as
|ψ〉AB ≡d−1∑i=0
√pi |i〉A |i〉B , where
probabilities pi are real, strictly positive, and normalized∑
i pi = 1.
{|i〉A} and {|i〉B} are orthonormal bases for systems A and B.[√pi]i∈{0,...,d−1} is the vector of Schmidt coefficients.
Schmidt rank d of |ψ〉AB is equal to the number of Schmidtcoefficients pi in its Schmidt decomposition and satisfies
d ≤ min {dim(HA), dim(HB)} .
State |ψ〉〈ψ|AB is entangled iff d ≥ 2.
Mark M. Wilde (LSU) 27 / 113
![Page 106: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/106.jpg)
Tool: Schmidt decomposition
Schmidt decomposition theorem
Given a two-party unit vector |ψ〉AB ∈ HA ⊗HB , we can express it as
|ψ〉AB ≡d−1∑i=0
√pi |i〉A |i〉B , where
probabilities pi are real, strictly positive, and normalized∑
i pi = 1.
{|i〉A} and {|i〉B} are orthonormal bases for systems A and B.[√pi]i∈{0,...,d−1} is the vector of Schmidt coefficients.
Schmidt rank d of |ψ〉AB is equal to the number of Schmidtcoefficients pi in its Schmidt decomposition and satisfies
d ≤ min {dim(HA), dim(HB)} .
State |ψ〉〈ψ|AB is entangled iff d ≥ 2.
Mark M. Wilde (LSU) 27 / 113
![Page 107: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/107.jpg)
Tool: Schmidt decomposition
Schmidt decomposition theorem
Given a two-party unit vector |ψ〉AB ∈ HA ⊗HB , we can express it as
|ψ〉AB ≡d−1∑i=0
√pi |i〉A |i〉B , where
probabilities pi are real, strictly positive, and normalized∑
i pi = 1.
{|i〉A} and {|i〉B} are orthonormal bases for systems A and B.[√pi]i∈{0,...,d−1} is the vector of Schmidt coefficients.
Schmidt rank d of |ψ〉AB is equal to the number of Schmidtcoefficients pi in its Schmidt decomposition and satisfies
d ≤ min {dim(HA), dim(HB)} .
State |ψ〉〈ψ|AB is entangled iff d ≥ 2.
Mark M. Wilde (LSU) 27 / 113
![Page 108: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/108.jpg)
Tool: Partial trace
The trace of a matrix X can be realized as
Tr{X} =∑i
〈i |X |i〉,
where {|i〉} is an orthonormal basis.
Partial trace of a matrix YAB acting on HA ⊗HB can be realized as
TrA{YAB} =∑i
(〈i |A ⊗ IB)YAB(|i〉A ⊗ IB),
where {|i〉A} is an orthonormal basis for HA and IB is the identitymatrix acting on HB .
Both trace and partial trace are linear operations.
Mark M. Wilde (LSU) 28 / 113
![Page 109: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/109.jpg)
Tool: Partial trace
The trace of a matrix X can be realized as
Tr{X} =∑i
〈i |X |i〉,
where {|i〉} is an orthonormal basis.
Partial trace of a matrix YAB acting on HA ⊗HB can be realized as
TrA{YAB} =∑i
(〈i |A ⊗ IB)YAB(|i〉A ⊗ IB),
where {|i〉A} is an orthonormal basis for HA and IB is the identitymatrix acting on HB .
Both trace and partial trace are linear operations.
Mark M. Wilde (LSU) 28 / 113
![Page 110: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/110.jpg)
Tool: Partial trace
The trace of a matrix X can be realized as
Tr{X} =∑i
〈i |X |i〉,
where {|i〉} is an orthonormal basis.
Partial trace of a matrix YAB acting on HA ⊗HB can be realized as
TrA{YAB} =∑i
(〈i |A ⊗ IB)YAB(|i〉A ⊗ IB),
where {|i〉A} is an orthonormal basis for HA and IB is the identitymatrix acting on HB .
Both trace and partial trace are linear operations.
Mark M. Wilde (LSU) 28 / 113
![Page 111: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/111.jpg)
Tool: Partial trace
The trace of a matrix X can be realized as
Tr{X} =∑i
〈i |X |i〉,
where {|i〉} is an orthonormal basis.
Partial trace of a matrix YAB acting on HA ⊗HB can be realized as
TrA{YAB} =∑i
(〈i |A ⊗ IB)YAB(|i〉A ⊗ IB),
where {|i〉A} is an orthonormal basis for HA and IB is the identitymatrix acting on HB .
Both trace and partial trace are linear operations.
Mark M. Wilde (LSU) 28 / 113
![Page 112: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/112.jpg)
Interpretation of partial trace
Suppose Alice and Bob possess quantum systems in the state ρAB .We calculate the density matrix for Alice’s system using partial trace:
ρA ≡ TrA{ρAB}.
We can then use ρA to predict the outcome of any experimentperformed on Alice’s system alone.
Partial trace generalizes marginalizing a probability distribution:
TrY
{∑x ,y
pX ,Y (x , y)|x〉〈x |X ⊗ |y〉〈y |Y
}=∑x ,y
pX ,Y (x , y)|x〉〈x |X Tr {|y〉〈y |Y }
=∑x
[∑y
pX ,Y (x , y)
]|x〉〈x |X =
∑x
pX (x)|x〉〈x |X ,
where pX (x) ≡∑
y pX ,Y (x , y).
Mark M. Wilde (LSU) 29 / 113
![Page 113: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/113.jpg)
Interpretation of partial trace
Suppose Alice and Bob possess quantum systems in the state ρAB .We calculate the density matrix for Alice’s system using partial trace:
ρA ≡ TrA{ρAB}.
We can then use ρA to predict the outcome of any experimentperformed on Alice’s system alone.
Partial trace generalizes marginalizing a probability distribution:
TrY
{∑x ,y
pX ,Y (x , y)|x〉〈x |X ⊗ |y〉〈y |Y
}=∑x ,y
pX ,Y (x , y)|x〉〈x |X Tr {|y〉〈y |Y }
=∑x
[∑y
pX ,Y (x , y)
]|x〉〈x |X =
∑x
pX (x)|x〉〈x |X ,
where pX (x) ≡∑
y pX ,Y (x , y).
Mark M. Wilde (LSU) 29 / 113
![Page 114: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/114.jpg)
Interpretation of partial trace
Suppose Alice and Bob possess quantum systems in the state ρAB .We calculate the density matrix for Alice’s system using partial trace:
ρA ≡ TrA{ρAB}.
We can then use ρA to predict the outcome of any experimentperformed on Alice’s system alone.
Partial trace generalizes marginalizing a probability distribution:
TrY
{∑x ,y
pX ,Y (x , y)|x〉〈x |X ⊗ |y〉〈y |Y
}=∑x ,y
pX ,Y (x , y)|x〉〈x |X Tr {|y〉〈y |Y }
=∑x
[∑y
pX ,Y (x , y)
]|x〉〈x |X =
∑x
pX (x)|x〉〈x |X ,
where pX (x) ≡∑
y pX ,Y (x , y).
Mark M. Wilde (LSU) 29 / 113
![Page 115: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/115.jpg)
Interpretation of partial trace
Suppose Alice and Bob possess quantum systems in the state ρAB .We calculate the density matrix for Alice’s system using partial trace:
ρA ≡ TrA{ρAB}.
We can then use ρA to predict the outcome of any experimentperformed on Alice’s system alone.
Partial trace generalizes marginalizing a probability distribution:
TrY
{∑x ,y
pX ,Y (x , y)|x〉〈x |X ⊗ |y〉〈y |Y
}=∑x ,y
pX ,Y (x , y)|x〉〈x |X Tr {|y〉〈y |Y }
=∑x
[∑y
pX ,Y (x , y)
]|x〉〈x |X =
∑x
pX (x)|x〉〈x |X ,
where pX (x) ≡∑
y pX ,Y (x , y).
Mark M. Wilde (LSU) 29 / 113
![Page 116: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/116.jpg)
Purification of quantum noise...
Artistic rendering of the notion of purification(Image courtesy of seaskylab at FreeDigitalPhotos.net)
Mark M. Wilde (LSU) 30 / 113
![Page 117: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/117.jpg)
Tool: Purification of quantum states
A purification of a state ρS on system S is a pure quantum state|ψ〉〈ψ|RS on systems R and S , such that
ρS = TrR{|ψ〉〈ψ|RS}.
Simple construction: take |ψ〉RS =∑
x
√p(x)|x〉R ⊗ |x〉S if ρS has
spectral decomposition∑
x p(x)|x〉〈x |S .
Two different states |ψ〉〈ψ|RS and |φ〉〈φ|RS purify ρS iff they arerelated by a unitary UR acting on the reference system. Necessity:
TrR{(UR ⊗ IS)|ψ〉〈ψ|RS(U†R ⊗ IS)} = TrR{(U†RUR ⊗ IS)|ψ〉〈ψ|RS}= TrR{|ψ〉〈ψ|RS}= ρS .
To prove sufficiency, use Schmidt decomposition.
Mark M. Wilde (LSU) 31 / 113
![Page 118: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/118.jpg)
Tool: Purification of quantum states
A purification of a state ρS on system S is a pure quantum state|ψ〉〈ψ|RS on systems R and S , such that
ρS = TrR{|ψ〉〈ψ|RS}.
Simple construction: take |ψ〉RS =∑
x
√p(x)|x〉R ⊗ |x〉S if ρS has
spectral decomposition∑
x p(x)|x〉〈x |S .
Two different states |ψ〉〈ψ|RS and |φ〉〈φ|RS purify ρS iff they arerelated by a unitary UR acting on the reference system. Necessity:
TrR{(UR ⊗ IS)|ψ〉〈ψ|RS(U†R ⊗ IS)} = TrR{(U†RUR ⊗ IS)|ψ〉〈ψ|RS}= TrR{|ψ〉〈ψ|RS}= ρS .
To prove sufficiency, use Schmidt decomposition.
Mark M. Wilde (LSU) 31 / 113
![Page 119: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/119.jpg)
Tool: Purification of quantum states
A purification of a state ρS on system S is a pure quantum state|ψ〉〈ψ|RS on systems R and S , such that
ρS = TrR{|ψ〉〈ψ|RS}.
Simple construction: take |ψ〉RS =∑
x
√p(x)|x〉R ⊗ |x〉S if ρS has
spectral decomposition∑
x p(x)|x〉〈x |S .
Two different states |ψ〉〈ψ|RS and |φ〉〈φ|RS purify ρS iff they arerelated by a unitary UR acting on the reference system. Necessity:
TrR{(UR ⊗ IS)|ψ〉〈ψ|RS(U†R ⊗ IS)} = TrR{(U†RUR ⊗ IS)|ψ〉〈ψ|RS}= TrR{|ψ〉〈ψ|RS}= ρS .
To prove sufficiency, use Schmidt decomposition.
Mark M. Wilde (LSU) 31 / 113
![Page 120: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/120.jpg)
Tool: Purification of quantum states
A purification of a state ρS on system S is a pure quantum state|ψ〉〈ψ|RS on systems R and S , such that
ρS = TrR{|ψ〉〈ψ|RS}.
Simple construction: take |ψ〉RS =∑
x
√p(x)|x〉R ⊗ |x〉S if ρS has
spectral decomposition∑
x p(x)|x〉〈x |S .
Two different states |ψ〉〈ψ|RS and |φ〉〈φ|RS purify ρS iff they arerelated by a unitary UR acting on the reference system. Necessity:
TrR{(UR ⊗ IS)|ψ〉〈ψ|RS(U†R ⊗ IS)} = TrR{(U†RUR ⊗ IS)|ψ〉〈ψ|RS}= TrR{|ψ〉〈ψ|RS}= ρS .
To prove sufficiency, use Schmidt decomposition.
Mark M. Wilde (LSU) 31 / 113
![Page 121: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/121.jpg)
Uses and interpretations of purification
The concept of purification is one of the most often used tools inquantum information theory.
This concept does not exist in classical information theory andrepresents a radical departure (i.e., in classical information theory it isnot possible to have a definite state of two systems such that thereduced systems are individually indefinite).
Physical interpretation: Noise or mixedness in a quantum state is dueto entanglement with an inaccessible reference / environment system.
Cryptographic interpretation: In the setting of quantum cryptography,we assume that an eavesdropper Eve has access to the fullpurification of a state ρAB that Alice and Bob share. This meansphysically that Eve has access to every other system in the universethat Alice and Bob do not have access to!
Advantage: only need to characterize Alice and Bob’s state in orderto understand what Eve has.
Mark M. Wilde (LSU) 32 / 113
![Page 122: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/122.jpg)
Uses and interpretations of purification
The concept of purification is one of the most often used tools inquantum information theory.
This concept does not exist in classical information theory andrepresents a radical departure (i.e., in classical information theory it isnot possible to have a definite state of two systems such that thereduced systems are individually indefinite).
Physical interpretation: Noise or mixedness in a quantum state is dueto entanglement with an inaccessible reference / environment system.
Cryptographic interpretation: In the setting of quantum cryptography,we assume that an eavesdropper Eve has access to the fullpurification of a state ρAB that Alice and Bob share. This meansphysically that Eve has access to every other system in the universethat Alice and Bob do not have access to!
Advantage: only need to characterize Alice and Bob’s state in orderto understand what Eve has.
Mark M. Wilde (LSU) 32 / 113
![Page 123: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/123.jpg)
Uses and interpretations of purification
The concept of purification is one of the most often used tools inquantum information theory.
This concept does not exist in classical information theory andrepresents a radical departure (i.e., in classical information theory it isnot possible to have a definite state of two systems such that thereduced systems are individually indefinite).
Physical interpretation: Noise or mixedness in a quantum state is dueto entanglement with an inaccessible reference / environment system.
Cryptographic interpretation: In the setting of quantum cryptography,we assume that an eavesdropper Eve has access to the fullpurification of a state ρAB that Alice and Bob share. This meansphysically that Eve has access to every other system in the universethat Alice and Bob do not have access to!
Advantage: only need to characterize Alice and Bob’s state in orderto understand what Eve has.
Mark M. Wilde (LSU) 32 / 113
![Page 124: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/124.jpg)
Uses and interpretations of purification
The concept of purification is one of the most often used tools inquantum information theory.
This concept does not exist in classical information theory andrepresents a radical departure (i.e., in classical information theory it isnot possible to have a definite state of two systems such that thereduced systems are individually indefinite).
Physical interpretation: Noise or mixedness in a quantum state is dueto entanglement with an inaccessible reference / environment system.
Cryptographic interpretation: In the setting of quantum cryptography,we assume that an eavesdropper Eve has access to the fullpurification of a state ρAB that Alice and Bob share. This meansphysically that Eve has access to every other system in the universethat Alice and Bob do not have access to!
Advantage: only need to characterize Alice and Bob’s state in orderto understand what Eve has.
Mark M. Wilde (LSU) 32 / 113
![Page 125: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/125.jpg)
Uses and interpretations of purification
The concept of purification is one of the most often used tools inquantum information theory.
This concept does not exist in classical information theory andrepresents a radical departure (i.e., in classical information theory it isnot possible to have a definite state of two systems such that thereduced systems are individually indefinite).
Physical interpretation: Noise or mixedness in a quantum state is dueto entanglement with an inaccessible reference / environment system.
Cryptographic interpretation: In the setting of quantum cryptography,we assume that an eavesdropper Eve has access to the fullpurification of a state ρAB that Alice and Bob share. This meansphysically that Eve has access to every other system in the universethat Alice and Bob do not have access to!
Advantage: only need to characterize Alice and Bob’s state in orderto understand what Eve has.
Mark M. Wilde (LSU) 32 / 113
![Page 126: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/126.jpg)
Uses and interpretations of purification
The concept of purification is one of the most often used tools inquantum information theory.
This concept does not exist in classical information theory andrepresents a radical departure (i.e., in classical information theory it isnot possible to have a definite state of two systems such that thereduced systems are individually indefinite).
Physical interpretation: Noise or mixedness in a quantum state is dueto entanglement with an inaccessible reference / environment system.
Cryptographic interpretation: In the setting of quantum cryptography,we assume that an eavesdropper Eve has access to the fullpurification of a state ρAB that Alice and Bob share. This meansphysically that Eve has access to every other system in the universethat Alice and Bob do not have access to!
Advantage: only need to characterize Alice and Bob’s state in orderto understand what Eve has.
Mark M. Wilde (LSU) 32 / 113
![Page 127: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/127.jpg)
Quantum channels
Mark M. Wilde (LSU) 33 / 113
![Page 128: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/128.jpg)
Classical channels
Classical channels model evolutions of classical systems.
What are the requirements that we make for classical channels?
1) They should be linear maps, which means they respect convexity.
2) They should take probability distributions to probabilitydistributions (i.e., they should output a legitimate state of a classicalsystem when a classical state is input).
These requirements imply that the evolution of a classical system isspecified by a conditional probability matrix N with entries pY |X (y |x),so that the input-output relationship of a classical channel is given by
pY = N pX ⇐⇒ pY (y) =∑x
pY |X (y |x)pX (x).
Mark M. Wilde (LSU) 34 / 113
![Page 129: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/129.jpg)
Classical channels
Classical channels model evolutions of classical systems.
What are the requirements that we make for classical channels?
1) They should be linear maps, which means they respect convexity.
2) They should take probability distributions to probabilitydistributions (i.e., they should output a legitimate state of a classicalsystem when a classical state is input).
These requirements imply that the evolution of a classical system isspecified by a conditional probability matrix N with entries pY |X (y |x),so that the input-output relationship of a classical channel is given by
pY = N pX ⇐⇒ pY (y) =∑x
pY |X (y |x)pX (x).
Mark M. Wilde (LSU) 34 / 113
![Page 130: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/130.jpg)
Classical channels
Classical channels model evolutions of classical systems.
What are the requirements that we make for classical channels?
1) They should be linear maps, which means they respect convexity.
2) They should take probability distributions to probabilitydistributions (i.e., they should output a legitimate state of a classicalsystem when a classical state is input).
These requirements imply that the evolution of a classical system isspecified by a conditional probability matrix N with entries pY |X (y |x),so that the input-output relationship of a classical channel is given by
pY = N pX ⇐⇒ pY (y) =∑x
pY |X (y |x)pX (x).
Mark M. Wilde (LSU) 34 / 113
![Page 131: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/131.jpg)
Classical channels
Classical channels model evolutions of classical systems.
What are the requirements that we make for classical channels?
1) They should be linear maps, which means they respect convexity.
2) They should take probability distributions to probabilitydistributions (i.e., they should output a legitimate state of a classicalsystem when a classical state is input).
These requirements imply that the evolution of a classical system isspecified by a conditional probability matrix N with entries pY |X (y |x),so that the input-output relationship of a classical channel is given by
pY = N pX ⇐⇒ pY (y) =∑x
pY |X (y |x)pX (x).
Mark M. Wilde (LSU) 34 / 113
![Page 132: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/132.jpg)
Classical channels
Classical channels model evolutions of classical systems.
What are the requirements that we make for classical channels?
1) They should be linear maps, which means they respect convexity.
2) They should take probability distributions to probabilitydistributions (i.e., they should output a legitimate state of a classicalsystem when a classical state is input).
These requirements imply that the evolution of a classical system isspecified by a conditional probability matrix N with entries pY |X (y |x),so that the input-output relationship of a classical channel is given by
pY = N pX ⇐⇒ pY (y) =∑x
pY |X (y |x)pX (x).
Mark M. Wilde (LSU) 34 / 113
![Page 133: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/133.jpg)
Classical channels
Classical channels model evolutions of classical systems.
What are the requirements that we make for classical channels?
1) They should be linear maps, which means they respect convexity.
2) They should take probability distributions to probabilitydistributions (i.e., they should output a legitimate state of a classicalsystem when a classical state is input).
These requirements imply that the evolution of a classical system isspecified by a conditional probability matrix N with entries pY |X (y |x),so that the input-output relationship of a classical channel is given by
pY = N pX ⇐⇒ pY (y) =∑x
pY |X (y |x)pX (x).
Mark M. Wilde (LSU) 34 / 113
![Page 134: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/134.jpg)
Quantum channels
Quantum channels model evolutions of quantum systems.
We make similar requirements:
A quantum channel N is a linear map acting on the space of(density) matrices:
N (pρ+ (1− p)σ) = pN (ρ) + (1− p)N (σ),
where p ∈ [0, 1] and ρ, σ ∈ D(H).
We demand that a quantum channel should take quantum states toquantum states.
This means that it should be trace (probability) preserving:
Tr{N (X )} = Tr{X}
for all X ∈ L(H) (linear operators, i.e., matrices).
Mark M. Wilde (LSU) 35 / 113
![Page 135: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/135.jpg)
Quantum channels
Quantum channels model evolutions of quantum systems.
We make similar requirements:
A quantum channel N is a linear map acting on the space of(density) matrices:
N (pρ+ (1− p)σ) = pN (ρ) + (1− p)N (σ),
where p ∈ [0, 1] and ρ, σ ∈ D(H).
We demand that a quantum channel should take quantum states toquantum states.
This means that it should be trace (probability) preserving:
Tr{N (X )} = Tr{X}
for all X ∈ L(H) (linear operators, i.e., matrices).
Mark M. Wilde (LSU) 35 / 113
![Page 136: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/136.jpg)
Quantum channels
Quantum channels model evolutions of quantum systems.
We make similar requirements:
A quantum channel N is a linear map acting on the space of(density) matrices:
N (pρ+ (1− p)σ) = pN (ρ) + (1− p)N (σ),
where p ∈ [0, 1] and ρ, σ ∈ D(H).
We demand that a quantum channel should take quantum states toquantum states.
This means that it should be trace (probability) preserving:
Tr{N (X )} = Tr{X}
for all X ∈ L(H) (linear operators, i.e., matrices).
Mark M. Wilde (LSU) 35 / 113
![Page 137: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/137.jpg)
Quantum channels
Quantum channels model evolutions of quantum systems.
We make similar requirements:
A quantum channel N is a linear map acting on the space of(density) matrices:
N (pρ+ (1− p)σ) = pN (ρ) + (1− p)N (σ),
where p ∈ [0, 1] and ρ, σ ∈ D(H).
We demand that a quantum channel should take quantum states toquantum states.
This means that it should be trace (probability) preserving:
Tr{N (X )} = Tr{X}
for all X ∈ L(H) (linear operators, i.e., matrices).
Mark M. Wilde (LSU) 35 / 113
![Page 138: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/138.jpg)
Quantum channels
Quantum channels model evolutions of quantum systems.
We make similar requirements:
A quantum channel N is a linear map acting on the space of(density) matrices:
N (pρ+ (1− p)σ) = pN (ρ) + (1− p)N (σ),
where p ∈ [0, 1] and ρ, σ ∈ D(H).
We demand that a quantum channel should take quantum states toquantum states.
This means that it should be trace (probability) preserving:
Tr{N (X )} = Tr{X}
for all X ∈ L(H) (linear operators, i.e., matrices).
Mark M. Wilde (LSU) 35 / 113
![Page 139: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/139.jpg)
Quantum channels
Quantum channels model evolutions of quantum systems.
We make similar requirements:
A quantum channel N is a linear map acting on the space of(density) matrices:
N (pρ+ (1− p)σ) = pN (ρ) + (1− p)N (σ),
where p ∈ [0, 1] and ρ, σ ∈ D(H).
We demand that a quantum channel should take quantum states toquantum states.
This means that it should be trace (probability) preserving:
Tr{N (X )} = Tr{X}
for all X ∈ L(H) (linear operators, i.e., matrices).
Mark M. Wilde (LSU) 35 / 113
![Page 140: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/140.jpg)
Complete positivity
Other requirement is complete positivity.
We can always expand XRS ∈ L(HR ⊗HS) as
XRS =∑i ,j
|i〉〈j |R ⊗ X i ,jS ,
and then define
(idR ⊗NS)(XRS) =∑i ,j
|i〉〈j |R ⊗NS
(X i ,jS
),
with the interpretation being that “nothing (identity channel)happens on system R while the channel N acts on system S .”
A quantum channel should also be completely positive:
(idR ⊗NS)(XRS) ≥ 0,
where idR denotes the identity channel acting on system R ofarbitrary size and XRS ∈ L(HR ⊗HS) is such that XRS ≥ 0.
Mark M. Wilde (LSU) 36 / 113
![Page 141: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/141.jpg)
Complete positivity
Other requirement is complete positivity.
We can always expand XRS ∈ L(HR ⊗HS) as
XRS =∑i ,j
|i〉〈j |R ⊗ X i ,jS ,
and then define
(idR ⊗NS)(XRS) =∑i ,j
|i〉〈j |R ⊗NS
(X i ,jS
),
with the interpretation being that “nothing (identity channel)happens on system R while the channel N acts on system S .”
A quantum channel should also be completely positive:
(idR ⊗NS)(XRS) ≥ 0,
where idR denotes the identity channel acting on system R ofarbitrary size and XRS ∈ L(HR ⊗HS) is such that XRS ≥ 0.
Mark M. Wilde (LSU) 36 / 113
![Page 142: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/142.jpg)
Complete positivity
Other requirement is complete positivity.
We can always expand XRS ∈ L(HR ⊗HS) as
XRS =∑i ,j
|i〉〈j |R ⊗ X i ,jS ,
and then define
(idR ⊗NS)(XRS) =∑i ,j
|i〉〈j |R ⊗NS
(X i ,jS
),
with the interpretation being that “nothing (identity channel)happens on system R while the channel N acts on system S .”
A quantum channel should also be completely positive:
(idR ⊗NS)(XRS) ≥ 0,
where idR denotes the identity channel acting on system R ofarbitrary size and XRS ∈ L(HR ⊗HS) is such that XRS ≥ 0.
Mark M. Wilde (LSU) 36 / 113
![Page 143: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/143.jpg)
Complete positivity
Other requirement is complete positivity.
We can always expand XRS ∈ L(HR ⊗HS) as
XRS =∑i ,j
|i〉〈j |R ⊗ X i ,jS ,
and then define
(idR ⊗NS)(XRS) =∑i ,j
|i〉〈j |R ⊗NS
(X i ,jS
),
with the interpretation being that “nothing (identity channel)happens on system R while the channel N acts on system S .”
A quantum channel should also be completely positive:
(idR ⊗NS)(XRS) ≥ 0,
where idR denotes the identity channel acting on system R ofarbitrary size and XRS ∈ L(HR ⊗HS) is such that XRS ≥ 0.
Mark M. Wilde (LSU) 36 / 113
![Page 144: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/144.jpg)
Quantum channels: completely positive, trace-preserving
A map N satisfying the requirements of linearity, trace preservation,and complete positivity takes all density matrices to density matricesand is called a quantum channel.
To check whether a given map is completely positive, it suffices tocheck whether
(idR ⊗NS)(|Φ〉〈Φ|RS) ≥ 0,
where
|Φ〉RS =1√d
∑i
|i〉R ⊗ |i〉S
and d = dim(HR) = dim(HS).
Interpretation: the state resulting from a channel acting on one shareof a maximally entangled state completely characterizes the channel.
Mark M. Wilde (LSU) 37 / 113
![Page 145: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/145.jpg)
Quantum channels: completely positive, trace-preserving
A map N satisfying the requirements of linearity, trace preservation,and complete positivity takes all density matrices to density matricesand is called a quantum channel.
To check whether a given map is completely positive, it suffices tocheck whether
(idR ⊗NS)(|Φ〉〈Φ|RS) ≥ 0,
where
|Φ〉RS =1√d
∑i
|i〉R ⊗ |i〉S
and d = dim(HR) = dim(HS).
Interpretation: the state resulting from a channel acting on one shareof a maximally entangled state completely characterizes the channel.
Mark M. Wilde (LSU) 37 / 113
![Page 146: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/146.jpg)
Quantum channels: completely positive, trace-preserving
A map N satisfying the requirements of linearity, trace preservation,and complete positivity takes all density matrices to density matricesand is called a quantum channel.
To check whether a given map is completely positive, it suffices tocheck whether
(idR ⊗NS)(|Φ〉〈Φ|RS) ≥ 0,
where
|Φ〉RS =1√d
∑i
|i〉R ⊗ |i〉S
and d = dim(HR) = dim(HS).
Interpretation: the state resulting from a channel acting on one shareof a maximally entangled state completely characterizes the channel.
Mark M. Wilde (LSU) 37 / 113
![Page 147: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/147.jpg)
Quantum channels: completely positive, trace-preserving
A map N satisfying the requirements of linearity, trace preservation,and complete positivity takes all density matrices to density matricesand is called a quantum channel.
To check whether a given map is completely positive, it suffices tocheck whether
(idR ⊗NS)(|Φ〉〈Φ|RS) ≥ 0,
where
|Φ〉RS =1√d
∑i
|i〉R ⊗ |i〉S
and d = dim(HR) = dim(HS).
Interpretation: the state resulting from a channel acting on one shareof a maximally entangled state completely characterizes the channel.
Mark M. Wilde (LSU) 37 / 113
![Page 148: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/148.jpg)
Choi-Kraus representation theorem
Structure theorem for quantum channels
Every quantum channel N can be written in the following form:
N (X ) =∑i
KiXK†i , (1)
where {Ki} is a set of Kraus operators, with the property that∑i
K †i Ki = I . (2)
The form given in (1) corresponds to complete positivity and the conditionin (2) to trace (probability) preservation. This decomposition is notunique, but one can find a minimal decomposition by taking a spectraldecomposition of (idR ⊗NS)(|Φ〉〈Φ|RS).
Mark M. Wilde (LSU) 38 / 113
![Page 149: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/149.jpg)
Examples of quantum channels
Quantum bit-flip channel for p ∈ [0, 1]:
ρ→ (1− p)ρ+ pXρX .
Quantum depolarizing channel for p ∈ [0, 1]:
ρ→ (1− p)ρ+ pπ,
where π ≡ I/d (maximally mixed state).
Quantum erasure channel for p ∈ [0, 1]:
ρ→ (1− p)ρ+ p|e〉〈e|,
where 〈e|ρ|e〉 = 0 for all inputs ρ.
Mark M. Wilde (LSU) 39 / 113
![Page 150: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/150.jpg)
Examples of quantum channels
Quantum bit-flip channel for p ∈ [0, 1]:
ρ→ (1− p)ρ+ pXρX .
Quantum depolarizing channel for p ∈ [0, 1]:
ρ→ (1− p)ρ+ pπ,
where π ≡ I/d (maximally mixed state).
Quantum erasure channel for p ∈ [0, 1]:
ρ→ (1− p)ρ+ p|e〉〈e|,
where 〈e|ρ|e〉 = 0 for all inputs ρ.
Mark M. Wilde (LSU) 39 / 113
![Page 151: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/151.jpg)
Examples of quantum channels
Quantum bit-flip channel for p ∈ [0, 1]:
ρ→ (1− p)ρ+ pXρX .
Quantum depolarizing channel for p ∈ [0, 1]:
ρ→ (1− p)ρ+ pπ,
where π ≡ I/d (maximally mixed state).
Quantum erasure channel for p ∈ [0, 1]:
ρ→ (1− p)ρ+ p|e〉〈e|,
where 〈e|ρ|e〉 = 0 for all inputs ρ.
Mark M. Wilde (LSU) 39 / 113
![Page 152: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/152.jpg)
Examples of quantum channels
Quantum bit-flip channel for p ∈ [0, 1]:
ρ→ (1− p)ρ+ pXρX .
Quantum depolarizing channel for p ∈ [0, 1]:
ρ→ (1− p)ρ+ pπ,
where π ≡ I/d (maximally mixed state).
Quantum erasure channel for p ∈ [0, 1]:
ρ→ (1− p)ρ+ p|e〉〈e|,
where 〈e|ρ|e〉 = 0 for all inputs ρ.
Mark M. Wilde (LSU) 39 / 113
![Page 153: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/153.jpg)
Unitary channels
If a channel has one Kraus operator (call it U), then it satisfiesU†U = I and is thus a unitary matrix.1
Unitary channels are ideal, reversible channels.
Instruction sequences for quantum algorithms (to be run on quantumcomputers) are composed of ideal, unitary channels.
So if a quantum channel has more than one Kraus operator (in aminimal decomposition), then it is non-unitary and irreversible.
1It could also be part of a unitary matrix, in which case it is called an “isometry.”Mark M. Wilde (LSU) 40 / 113
![Page 154: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/154.jpg)
Unitary channels
If a channel has one Kraus operator (call it U), then it satisfiesU†U = I and is thus a unitary matrix.1
Unitary channels are ideal, reversible channels.
Instruction sequences for quantum algorithms (to be run on quantumcomputers) are composed of ideal, unitary channels.
So if a quantum channel has more than one Kraus operator (in aminimal decomposition), then it is non-unitary and irreversible.
1It could also be part of a unitary matrix, in which case it is called an “isometry.”Mark M. Wilde (LSU) 40 / 113
![Page 155: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/155.jpg)
Unitary channels
If a channel has one Kraus operator (call it U), then it satisfiesU†U = I and is thus a unitary matrix.1
Unitary channels are ideal, reversible channels.
Instruction sequences for quantum algorithms (to be run on quantumcomputers) are composed of ideal, unitary channels.
So if a quantum channel has more than one Kraus operator (in aminimal decomposition), then it is non-unitary and irreversible.
1It could also be part of a unitary matrix, in which case it is called an “isometry.”Mark M. Wilde (LSU) 40 / 113
![Page 156: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/156.jpg)
Unitary channels
If a channel has one Kraus operator (call it U), then it satisfiesU†U = I and is thus a unitary matrix.1
Unitary channels are ideal, reversible channels.
Instruction sequences for quantum algorithms (to be run on quantumcomputers) are composed of ideal, unitary channels.
So if a quantum channel has more than one Kraus operator (in aminimal decomposition), then it is non-unitary and irreversible.
1It could also be part of a unitary matrix, in which case it is called an “isometry.”Mark M. Wilde (LSU) 40 / 113
![Page 157: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/157.jpg)
Unitary channels
If a channel has one Kraus operator (call it U), then it satisfiesU†U = I and is thus a unitary matrix.1
Unitary channels are ideal, reversible channels.
Instruction sequences for quantum algorithms (to be run on quantumcomputers) are composed of ideal, unitary channels.
So if a quantum channel has more than one Kraus operator (in aminimal decomposition), then it is non-unitary and irreversible.
1It could also be part of a unitary matrix, in which case it is called an “isometry.”Mark M. Wilde (LSU) 40 / 113
![Page 158: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/158.jpg)
Preparation channels
Preparation channels take classical systems as input and producequantum systems as output.
A preparation channel P has the following form:
P(ρ) =∑x
〈x |ρ|x〉σx ,
where {|x〉} is an orthonormal basis and {σx} is a set of states.
Inputting the classical state |x〉〈x | leads to quantum output σx , i.e.,it is just the map
x → σx ,
where x is a classical letter. Sometimes called “cq” channel, short for“classical-to-quantum” channel.
Mark M. Wilde (LSU) 41 / 113
![Page 159: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/159.jpg)
Preparation channels
Preparation channels take classical systems as input and producequantum systems as output.
A preparation channel P has the following form:
P(ρ) =∑x
〈x |ρ|x〉σx ,
where {|x〉} is an orthonormal basis and {σx} is a set of states.
Inputting the classical state |x〉〈x | leads to quantum output σx , i.e.,it is just the map
x → σx ,
where x is a classical letter. Sometimes called “cq” channel, short for“classical-to-quantum” channel.
Mark M. Wilde (LSU) 41 / 113
![Page 160: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/160.jpg)
Preparation channels
Preparation channels take classical systems as input and producequantum systems as output.
A preparation channel P has the following form:
P(ρ) =∑x
〈x |ρ|x〉σx ,
where {|x〉} is an orthonormal basis and {σx} is a set of states.
Inputting the classical state |x〉〈x | leads to quantum output σx , i.e.,it is just the map
x → σx ,
where x is a classical letter. Sometimes called “cq” channel, short for“classical-to-quantum” channel.
Mark M. Wilde (LSU) 41 / 113
![Page 161: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/161.jpg)
Preparation channels
Preparation channels take classical systems as input and producequantum systems as output.
A preparation channel P has the following form:
P(ρ) =∑x
〈x |ρ|x〉σx ,
where {|x〉} is an orthonormal basis and {σx} is a set of states.
Inputting the classical state |x〉〈x | leads to quantum output σx , i.e.,it is just the map
x → σx ,
where x is a classical letter. Sometimes called “cq” channel, short for“classical-to-quantum” channel.
Mark M. Wilde (LSU) 41 / 113
![Page 162: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/162.jpg)
Measurement channels
Measurement channels take quantum systems as input and produceclassical systems as output.
A measurement channel M has the following form:
M(ρ) =∑x
Tr{Mxρ}|x〉〈x |,
where Mx ≥ 0 for all x and∑
x Mx = I .
Can also interpret a measurement channel as returning the classicalvalue x with probability Tr{Mxρ}.
We depict them as
Mark M. Wilde (LSU) 42 / 113
![Page 163: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/163.jpg)
Measurement channels
Measurement channels take quantum systems as input and produceclassical systems as output.
A measurement channel M has the following form:
M(ρ) =∑x
Tr{Mxρ}|x〉〈x |,
where Mx ≥ 0 for all x and∑
x Mx = I .
Can also interpret a measurement channel as returning the classicalvalue x with probability Tr{Mxρ}.
We depict them as
Mark M. Wilde (LSU) 42 / 113
![Page 164: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/164.jpg)
Measurement channels
Measurement channels take quantum systems as input and produceclassical systems as output.
A measurement channel M has the following form:
M(ρ) =∑x
Tr{Mxρ}|x〉〈x |,
where Mx ≥ 0 for all x and∑
x Mx = I .
Can also interpret a measurement channel as returning the classicalvalue x with probability Tr{Mxρ}.
We depict them as
Mark M. Wilde (LSU) 42 / 113
![Page 165: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/165.jpg)
Measurement channels
Measurement channels take quantum systems as input and produceclassical systems as output.
A measurement channel M has the following form:
M(ρ) =∑x
Tr{Mxρ}|x〉〈x |,
where Mx ≥ 0 for all x and∑
x Mx = I .
Can also interpret a measurement channel as returning the classicalvalue x with probability Tr{Mxρ}.
We depict them as
Mark M. Wilde (LSU) 42 / 113
![Page 166: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/166.jpg)
Measurement channels
Measurement channels take quantum systems as input and produceclassical systems as output.
A measurement channel M has the following form:
M(ρ) =∑x
Tr{Mxρ}|x〉〈x |,
where Mx ≥ 0 for all x and∑
x Mx = I .
Can also interpret a measurement channel as returning the classicalvalue x with probability Tr{Mxρ}.
We depict them as
Mark M. Wilde (LSU) 42 / 113
![Page 167: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/167.jpg)
“Measuring an operator”
Let G be a Hermitian operator with spectral decomposition
G =∑x
µxΠx ,
where µx are real eigenvalues and Πx are projections ontocorresponding eigensubspaces.
We say that an experimenter “measures an operator G” byperforming the following measurement channel:
ρ→∑x
Tr{Πxρ}|x〉〈x |,
where {|x〉} is an orthonormal basis.
Mark M. Wilde (LSU) 43 / 113
![Page 168: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/168.jpg)
“Measuring an operator”
Let G be a Hermitian operator with spectral decomposition
G =∑x
µxΠx ,
where µx are real eigenvalues and Πx are projections ontocorresponding eigensubspaces.
We say that an experimenter “measures an operator G” byperforming the following measurement channel:
ρ→∑x
Tr{Πxρ}|x〉〈x |,
where {|x〉} is an orthonormal basis.
Mark M. Wilde (LSU) 43 / 113
![Page 169: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/169.jpg)
“Measuring an operator”
Let G be a Hermitian operator with spectral decomposition
G =∑x
µxΠx ,
where µx are real eigenvalues and Πx are projections ontocorresponding eigensubspaces.
We say that an experimenter “measures an operator G” byperforming the following measurement channel:
ρ→∑x
Tr{Πxρ}|x〉〈x |,
where {|x〉} is an orthonormal basis.
Mark M. Wilde (LSU) 43 / 113
![Page 170: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/170.jpg)
Entanglement-breaking channels
An entanglement-breaking channel N is defined such that for everyinput state ρRS , the output
(idR ⊗NS)(ρRS)
is a separable state.
To determine whether a given channel is entanglement-breaking, itsuffices to check whether the following state is separable:
(idR ⊗NS)(|Φ〉〈Φ|RS).
Mark M. Wilde (LSU) 44 / 113
![Page 171: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/171.jpg)
Entanglement-breaking channels
An entanglement-breaking channel N is defined such that for everyinput state ρRS , the output
(idR ⊗NS)(ρRS)
is a separable state.
To determine whether a given channel is entanglement-breaking, itsuffices to check whether the following state is separable:
(idR ⊗NS)(|Φ〉〈Φ|RS).
Mark M. Wilde (LSU) 44 / 113
![Page 172: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/172.jpg)
Entanglement-breaking channels
An entanglement-breaking channel N is defined such that for everyinput state ρRS , the output
(idR ⊗NS)(ρRS)
is a separable state.
To determine whether a given channel is entanglement-breaking, itsuffices to check whether the following state is separable:
(idR ⊗NS)(|Φ〉〈Φ|RS).
Mark M. Wilde (LSU) 44 / 113
![Page 173: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/173.jpg)
Entanglement-breaking channels
Every entanglement-breaking (EB) channel N can be written as acomposition of a measurement M followed by a preparation P:
N = P ◦M.
Thus, internally, every EB channel transforms a quantum system to aclassical one and then back: q → c → q. In this sense, such channelsare one step up from classical channels and inherit some properties ofclassical channels.
Mark M. Wilde (LSU) 45 / 113
![Page 174: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/174.jpg)
Entanglement-breaking channels
Every entanglement-breaking (EB) channel N can be written as acomposition of a measurement M followed by a preparation P:
N = P ◦M.
Thus, internally, every EB channel transforms a quantum system to aclassical one and then back: q → c → q. In this sense, such channelsare one step up from classical channels and inherit some properties ofclassical channels.
Mark M. Wilde (LSU) 45 / 113
![Page 175: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/175.jpg)
Entanglement-breaking channels
Every entanglement-breaking (EB) channel N can be written as acomposition of a measurement M followed by a preparation P:
N = P ◦M.
Thus, internally, every EB channel transforms a quantum system to aclassical one and then back: q → c → q. In this sense, such channelsare one step up from classical channels and inherit some properties ofclassical channels.
Mark M. Wilde (LSU) 45 / 113
![Page 176: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/176.jpg)
Purifications of quantum channels
Recall that we can purify quantum states and understand noise asarising due to entanglement with an inaccessible reference system.
We can also purify quantum channels and understand a noisy processas arising from a unitary interaction with an inaccessible environment.
Stinespring’s theorem
For every quantum channel NA→B , there exists a pure state |0〉〈0|E and aunitary matrix UAE→BE ′ , acting on input systems A and E and producingoutput systems B and E ′, such that
NA→B(ρA) = TrE ′{UAE→BE ′(ρA ⊗ |0〉〈0|E )(UAE→BE ′)†}.
Mark M. Wilde (LSU) 46 / 113
![Page 177: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/177.jpg)
Purifications of quantum channels
Recall that we can purify quantum states and understand noise asarising due to entanglement with an inaccessible reference system.
We can also purify quantum channels and understand a noisy processas arising from a unitary interaction with an inaccessible environment.
Stinespring’s theorem
For every quantum channel NA→B , there exists a pure state |0〉〈0|E and aunitary matrix UAE→BE ′ , acting on input systems A and E and producingoutput systems B and E ′, such that
NA→B(ρA) = TrE ′{UAE→BE ′(ρA ⊗ |0〉〈0|E )(UAE→BE ′)†}.
Mark M. Wilde (LSU) 46 / 113
![Page 178: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/178.jpg)
Purifications of quantum channels
Recall that we can purify quantum states and understand noise asarising due to entanglement with an inaccessible reference system.
We can also purify quantum channels and understand a noisy processas arising from a unitary interaction with an inaccessible environment.
Stinespring’s theorem
For every quantum channel NA→B , there exists a pure state |0〉〈0|E and aunitary matrix UAE→BE ′ , acting on input systems A and E and producingoutput systems B and E ′, such that
NA→B(ρA) = TrE ′{UAE→BE ′(ρA ⊗ |0〉〈0|E )(UAE→BE ′)†}.
Mark M. Wilde (LSU) 46 / 113
![Page 179: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/179.jpg)
Purifications of quantum channels
Recall that we can purify quantum states and understand noise asarising due to entanglement with an inaccessible reference system.
We can also purify quantum channels and understand a noisy processas arising from a unitary interaction with an inaccessible environment.
Stinespring’s theorem
For every quantum channel NA→B , there exists a pure state |0〉〈0|E and aunitary matrix UAE→BE ′ , acting on input systems A and E and producingoutput systems B and E ′, such that
NA→B(ρA) = TrE ′{UAE→BE ′(ρA ⊗ |0〉〈0|E )(UAE→BE ′)†}.
Mark M. Wilde (LSU) 46 / 113
![Page 180: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/180.jpg)
Construction of a unitary extension
Standard construction of a unitary extension of a quantum channel:Given Kraus operators {Ki} for N such that N (ρ) =
∑i KiρK
†i , take
V =∑i
Ki ⊗ |i〉E ′〈0|E .
V †V = I , so we can fill in other columns such that matrix is unitary(call the result U).
Then
U(ρA ⊗ |0〉〈0|E )U† =∑i ,j
KiρK†j ⊗ |i〉〈j |E ′ ,
and TrE ′{U(ρA ⊗ |0〉〈0|E )U†} = TrE ′
∑i ,j
KiρK†j ⊗ |i〉〈j |E ′
=∑i
KiρK†i = N (ρ).
Mark M. Wilde (LSU) 47 / 113
![Page 181: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/181.jpg)
Construction of a unitary extension
Standard construction of a unitary extension of a quantum channel:Given Kraus operators {Ki} for N such that N (ρ) =
∑i KiρK
†i , take
V =∑i
Ki ⊗ |i〉E ′〈0|E .
V †V = I , so we can fill in other columns such that matrix is unitary(call the result U).
Then
U(ρA ⊗ |0〉〈0|E )U† =∑i ,j
KiρK†j ⊗ |i〉〈j |E ′ ,
and TrE ′{U(ρA ⊗ |0〉〈0|E )U†} = TrE ′
∑i ,j
KiρK†j ⊗ |i〉〈j |E ′
=∑i
KiρK†i = N (ρ).
Mark M. Wilde (LSU) 47 / 113
![Page 182: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/182.jpg)
Construction of a unitary extension
Standard construction of a unitary extension of a quantum channel:Given Kraus operators {Ki} for N such that N (ρ) =
∑i KiρK
†i , take
V =∑i
Ki ⊗ |i〉E ′〈0|E .
V †V = I , so we can fill in other columns such that matrix is unitary(call the result U).
Then
U(ρA ⊗ |0〉〈0|E )U† =∑i ,j
KiρK†j ⊗ |i〉〈j |E ′ ,
and TrE ′{U(ρA ⊗ |0〉〈0|E )U†} = TrE ′
∑i ,j
KiρK†j ⊗ |i〉〈j |E ′
=∑i
KiρK†i = N (ρ).
Mark M. Wilde (LSU) 47 / 113
![Page 183: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/183.jpg)
Construction of a unitary extension
Standard construction of a unitary extension of a quantum channel:Given Kraus operators {Ki} for N such that N (ρ) =
∑i KiρK
†i , take
V =∑i
Ki ⊗ |i〉E ′〈0|E .
V †V = I , so we can fill in other columns such that matrix is unitary(call the result U).
Then
U(ρA ⊗ |0〉〈0|E )U† =∑i ,j
KiρK†j ⊗ |i〉〈j |E ′ ,
and TrE ′{U(ρA ⊗ |0〉〈0|E )U†} = TrE ′
∑i ,j
KiρK†j ⊗ |i〉〈j |E ′
=∑i
KiρK†i = N (ρ).
Mark M. Wilde (LSU) 47 / 113
![Page 184: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/184.jpg)
Summary of quantum states and channels
Every quantum state is a positive, semi-definite matrix with traceequal to one.
Quantum states of multiple systems can be separable or entangled.
Quantum states can be purified (this notion does not exist in classicalinformation theory).
Quantum channels are completely positive, trace-preserving maps.
Preparation channels take classical systems to quantum systems, andmeasurement channels take quantum systems to classical systems.
Quantum channels can also be purified (i.e., every quantum channelcan be realized by a unitary interaction with an environment, followedby partial trace). This notion also does not exist in classicalinformation theory.
Mark M. Wilde (LSU) 48 / 113
![Page 185: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/185.jpg)
Summary of quantum states and channels
Every quantum state is a positive, semi-definite matrix with traceequal to one.
Quantum states of multiple systems can be separable or entangled.
Quantum states can be purified (this notion does not exist in classicalinformation theory).
Quantum channels are completely positive, trace-preserving maps.
Preparation channels take classical systems to quantum systems, andmeasurement channels take quantum systems to classical systems.
Quantum channels can also be purified (i.e., every quantum channelcan be realized by a unitary interaction with an environment, followedby partial trace). This notion also does not exist in classicalinformation theory.
Mark M. Wilde (LSU) 48 / 113
![Page 186: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/186.jpg)
Summary of quantum states and channels
Every quantum state is a positive, semi-definite matrix with traceequal to one.
Quantum states of multiple systems can be separable or entangled.
Quantum states can be purified (this notion does not exist in classicalinformation theory).
Quantum channels are completely positive, trace-preserving maps.
Preparation channels take classical systems to quantum systems, andmeasurement channels take quantum systems to classical systems.
Quantum channels can also be purified (i.e., every quantum channelcan be realized by a unitary interaction with an environment, followedby partial trace). This notion also does not exist in classicalinformation theory.
Mark M. Wilde (LSU) 48 / 113
![Page 187: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/187.jpg)
Summary of quantum states and channels
Every quantum state is a positive, semi-definite matrix with traceequal to one.
Quantum states of multiple systems can be separable or entangled.
Quantum states can be purified (this notion does not exist in classicalinformation theory).
Quantum channels are completely positive, trace-preserving maps.
Preparation channels take classical systems to quantum systems, andmeasurement channels take quantum systems to classical systems.
Quantum channels can also be purified (i.e., every quantum channelcan be realized by a unitary interaction with an environment, followedby partial trace). This notion also does not exist in classicalinformation theory.
Mark M. Wilde (LSU) 48 / 113
![Page 188: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/188.jpg)
Summary of quantum states and channels
Every quantum state is a positive, semi-definite matrix with traceequal to one.
Quantum states of multiple systems can be separable or entangled.
Quantum states can be purified (this notion does not exist in classicalinformation theory).
Quantum channels are completely positive, trace-preserving maps.
Preparation channels take classical systems to quantum systems, andmeasurement channels take quantum systems to classical systems.
Quantum channels can also be purified (i.e., every quantum channelcan be realized by a unitary interaction with an environment, followedby partial trace). This notion also does not exist in classicalinformation theory.
Mark M. Wilde (LSU) 48 / 113
![Page 189: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/189.jpg)
Summary of quantum states and channels
Every quantum state is a positive, semi-definite matrix with traceequal to one.
Quantum states of multiple systems can be separable or entangled.
Quantum states can be purified (this notion does not exist in classicalinformation theory).
Quantum channels are completely positive, trace-preserving maps.
Preparation channels take classical systems to quantum systems, andmeasurement channels take quantum systems to classical systems.
Quantum channels can also be purified (i.e., every quantum channelcan be realized by a unitary interaction with an environment, followedby partial trace). This notion also does not exist in classicalinformation theory.
Mark M. Wilde (LSU) 48 / 113
![Page 190: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/190.jpg)
Summary of quantum states and channels
Every quantum state is a positive, semi-definite matrix with traceequal to one.
Quantum states of multiple systems can be separable or entangled.
Quantum states can be purified (this notion does not exist in classicalinformation theory).
Quantum channels are completely positive, trace-preserving maps.
Preparation channels take classical systems to quantum systems, andmeasurement channels take quantum systems to classical systems.
Quantum channels can also be purified (i.e., every quantum channelcan be realized by a unitary interaction with an environment, followedby partial trace). This notion also does not exist in classicalinformation theory.
Mark M. Wilde (LSU) 48 / 113
![Page 191: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/191.jpg)
Fundamental protocols
Mark M. Wilde (LSU) 49 / 113
![Page 192: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/192.jpg)
Bell experiment / CHSH game
How is quantum information different from classical information?
One way to answer this question is to devise operational tasks forwhich a quantum strategy outperforms a classical one.
The most famous is the Bell experiment / CHSH game.2
The game involves two spatially separated parties (the players Aliceand Bob) and a referee.
2A “loop-hole free” implementation of this experiment was conducted in 2015 (seearXiv:1508.05949).
Mark M. Wilde (LSU) 50 / 113
![Page 193: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/193.jpg)
Bell experiment / CHSH game
How is quantum information different from classical information?
One way to answer this question is to devise operational tasks forwhich a quantum strategy outperforms a classical one.
The most famous is the Bell experiment / CHSH game.2
The game involves two spatially separated parties (the players Aliceand Bob) and a referee.
2A “loop-hole free” implementation of this experiment was conducted in 2015 (seearXiv:1508.05949).
Mark M. Wilde (LSU) 50 / 113
![Page 194: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/194.jpg)
Bell experiment / CHSH game
How is quantum information different from classical information?
One way to answer this question is to devise operational tasks forwhich a quantum strategy outperforms a classical one.
The most famous is the Bell experiment / CHSH game.2
The game involves two spatially separated parties (the players Aliceand Bob) and a referee.
2A “loop-hole free” implementation of this experiment was conducted in 2015 (seearXiv:1508.05949).
Mark M. Wilde (LSU) 50 / 113
![Page 195: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/195.jpg)
Bell experiment / CHSH game
How is quantum information different from classical information?
One way to answer this question is to devise operational tasks forwhich a quantum strategy outperforms a classical one.
The most famous is the Bell experiment / CHSH game.2
The game involves two spatially separated parties (the players Aliceand Bob) and a referee.
2A “loop-hole free” implementation of this experiment was conducted in 2015 (seearXiv:1508.05949).
Mark M. Wilde (LSU) 50 / 113
![Page 196: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/196.jpg)
Bell experiment / CHSH game
How is quantum information different from classical information?
One way to answer this question is to devise operational tasks forwhich a quantum strategy outperforms a classical one.
The most famous is the Bell experiment / CHSH game.2
The game involves two spatially separated parties (the players Aliceand Bob) and a referee.
2A “loop-hole free” implementation of this experiment was conducted in 2015 (seearXiv:1508.05949).
Mark M. Wilde (LSU) 50 / 113
![Page 197: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/197.jpg)
Bell experiment / CHSH game
Game begins with referee randomly picking bits x and y .
Referee sends x and y to Alice and Bob, respectively.
Alice replies with a bit a and Bob with a bit b.
They win if and only if a⊕ b = x ∧ y .
Mark M. Wilde (LSU) 51 / 113
![Page 198: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/198.jpg)
Bell experiment / CHSH game
Game begins with referee randomly picking bits x and y .
Referee sends x and y to Alice and Bob, respectively.
Alice replies with a bit a and Bob with a bit b.
They win if and only if a⊕ b = x ∧ y .
Mark M. Wilde (LSU) 51 / 113
![Page 199: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/199.jpg)
Bell experiment / CHSH game
Game begins with referee randomly picking bits x and y .
Referee sends x and y to Alice and Bob, respectively.
Alice replies with a bit a and Bob with a bit b.
They win if and only if a⊕ b = x ∧ y .
Mark M. Wilde (LSU) 51 / 113
![Page 200: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/200.jpg)
Bell experiment / CHSH game
Game begins with referee randomly picking bits x and y .
Referee sends x and y to Alice and Bob, respectively.
Alice replies with a bit a and Bob with a bit b.
They win if and only if a⊕ b = x ∧ y .
Mark M. Wilde (LSU) 51 / 113
![Page 201: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/201.jpg)
Bell experiment / CHSH game
Game begins with referee randomly picking bits x and y .
Referee sends x and y to Alice and Bob, respectively.
Alice replies with a bit a and Bob with a bit b.
They win if and only if a⊕ b = x ∧ y .
Mark M. Wilde (LSU) 51 / 113
![Page 202: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/202.jpg)
Classical strategies
The most general classical strategy allows for Alice and Bob topossess shared randomness before the game begins.
However, can show that shared randomness does not help them win.
Thus, to compute the winning probability with classical strategies, itsuffices to consider deterministic classical strategies.
Mark M. Wilde (LSU) 52 / 113
![Page 203: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/203.jpg)
Classical strategies
The most general classical strategy allows for Alice and Bob topossess shared randomness before the game begins.
However, can show that shared randomness does not help them win.
Thus, to compute the winning probability with classical strategies, itsuffices to consider deterministic classical strategies.
Mark M. Wilde (LSU) 52 / 113
![Page 204: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/204.jpg)
Classical strategies
The most general classical strategy allows for Alice and Bob topossess shared randomness before the game begins.
However, can show that shared randomness does not help them win.
Thus, to compute the winning probability with classical strategies, itsuffices to consider deterministic classical strategies.
Mark M. Wilde (LSU) 52 / 113
![Page 205: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/205.jpg)
Classical strategies
The most general classical strategy allows for Alice and Bob topossess shared randomness before the game begins.
However, can show that shared randomness does not help them win.
Thus, to compute the winning probability with classical strategies, itsuffices to consider deterministic classical strategies.
Mark M. Wilde (LSU) 52 / 113
![Page 206: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/206.jpg)
Deterministic classical strategies
General deterministic strategy: x → ax for Alice and y → by for Bob.
The following table presents the winning conditions for the fourdifferent values of x and y using this deterministic strategy:
x y x ∧ y = ax ⊕ by
0 0 0 = a0 ⊕ b0
0 1 0 = a0 ⊕ b1
1 0 0 = a1 ⊕ b0
1 1 1 = a1 ⊕ b1
They cannot always win. (If they could, there would be acontradiction, because adding up 3rd column gives 1 while adding up4th column gives 0.)
The best they can do is to win only 3/4 = 0.75 of the time!
Strategy achieving this: Alice and Bob each always report back zero.
Mark M. Wilde (LSU) 53 / 113
![Page 207: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/207.jpg)
Deterministic classical strategies
General deterministic strategy: x → ax for Alice and y → by for Bob.
The following table presents the winning conditions for the fourdifferent values of x and y using this deterministic strategy:
x y x ∧ y = ax ⊕ by
0 0 0 = a0 ⊕ b0
0 1 0 = a0 ⊕ b1
1 0 0 = a1 ⊕ b0
1 1 1 = a1 ⊕ b1
They cannot always win. (If they could, there would be acontradiction, because adding up 3rd column gives 1 while adding up4th column gives 0.)
The best they can do is to win only 3/4 = 0.75 of the time!
Strategy achieving this: Alice and Bob each always report back zero.
Mark M. Wilde (LSU) 53 / 113
![Page 208: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/208.jpg)
Deterministic classical strategies
General deterministic strategy: x → ax for Alice and y → by for Bob.
The following table presents the winning conditions for the fourdifferent values of x and y using this deterministic strategy:
x y x ∧ y = ax ⊕ by
0 0 0 = a0 ⊕ b0
0 1 0 = a0 ⊕ b1
1 0 0 = a1 ⊕ b0
1 1 1 = a1 ⊕ b1
They cannot always win. (If they could, there would be acontradiction, because adding up 3rd column gives 1 while adding up4th column gives 0.)
The best they can do is to win only 3/4 = 0.75 of the time!
Strategy achieving this: Alice and Bob each always report back zero.
Mark M. Wilde (LSU) 53 / 113
![Page 209: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/209.jpg)
Deterministic classical strategies
General deterministic strategy: x → ax for Alice and y → by for Bob.
The following table presents the winning conditions for the fourdifferent values of x and y using this deterministic strategy:
x y x ∧ y = ax ⊕ by
0 0 0 = a0 ⊕ b0
0 1 0 = a0 ⊕ b1
1 0 0 = a1 ⊕ b0
1 1 1 = a1 ⊕ b1
They cannot always win. (If they could, there would be acontradiction, because adding up 3rd column gives 1 while adding up4th column gives 0.)
The best they can do is to win only 3/4 = 0.75 of the time!
Strategy achieving this: Alice and Bob each always report back zero.
Mark M. Wilde (LSU) 53 / 113
![Page 210: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/210.jpg)
Deterministic classical strategies
General deterministic strategy: x → ax for Alice and y → by for Bob.
The following table presents the winning conditions for the fourdifferent values of x and y using this deterministic strategy:
x y x ∧ y = ax ⊕ by
0 0 0 = a0 ⊕ b0
0 1 0 = a0 ⊕ b1
1 0 0 = a1 ⊕ b0
1 1 1 = a1 ⊕ b1
They cannot always win. (If they could, there would be acontradiction, because adding up 3rd column gives 1 while adding up4th column gives 0.)
The best they can do is to win only 3/4 = 0.75 of the time!
Strategy achieving this: Alice and Bob each always report back zero.
Mark M. Wilde (LSU) 53 / 113
![Page 211: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/211.jpg)
Deterministic classical strategies
General deterministic strategy: x → ax for Alice and y → by for Bob.
The following table presents the winning conditions for the fourdifferent values of x and y using this deterministic strategy:
x y x ∧ y = ax ⊕ by
0 0 0 = a0 ⊕ b0
0 1 0 = a0 ⊕ b1
1 0 0 = a1 ⊕ b0
1 1 1 = a1 ⊕ b1
They cannot always win. (If they could, there would be acontradiction, because adding up 3rd column gives 1 while adding up4th column gives 0.)
The best they can do is to win only 3/4 = 0.75 of the time!
Strategy achieving this: Alice and Bob each always report back zero.
Mark M. Wilde (LSU) 53 / 113
![Page 212: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/212.jpg)
Quantum strategy
Allow Alice and Bob to share two qubits in the state |Φ〉〈Φ|AB beforethe game starts.
If Alice receives x = 0, then she performs a measurement of Z . If shereceives x = 1, then she performs a measurement of X . In each case,she reports the outcome as a.
If Bob receives y = 0, then he performs a measurement of(X + Z )/
√2. If he receives y = 1, then he performs a measurement
of (Z − X )/√
2. In each case, he reports the outcome as b.
This quantum strategy has a winning probability ofcos2(π/8) ≈ 0.85 > 0.75 and thus represents a significant separationbetween classical and quantum information theory.
Mark M. Wilde (LSU) 54 / 113
![Page 213: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/213.jpg)
Quantum strategy
Allow Alice and Bob to share two qubits in the state |Φ〉〈Φ|AB beforethe game starts.
If Alice receives x = 0, then she performs a measurement of Z . If shereceives x = 1, then she performs a measurement of X . In each case,she reports the outcome as a.
If Bob receives y = 0, then he performs a measurement of(X + Z )/
√2. If he receives y = 1, then he performs a measurement
of (Z − X )/√
2. In each case, he reports the outcome as b.
This quantum strategy has a winning probability ofcos2(π/8) ≈ 0.85 > 0.75 and thus represents a significant separationbetween classical and quantum information theory.
Mark M. Wilde (LSU) 54 / 113
![Page 214: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/214.jpg)
Quantum strategy
Allow Alice and Bob to share two qubits in the state |Φ〉〈Φ|AB beforethe game starts.
If Alice receives x = 0, then she performs a measurement of Z . If shereceives x = 1, then she performs a measurement of X . In each case,she reports the outcome as a.
If Bob receives y = 0, then he performs a measurement of(X + Z )/
√2. If he receives y = 1, then he performs a measurement
of (Z − X )/√
2. In each case, he reports the outcome as b.
This quantum strategy has a winning probability ofcos2(π/8) ≈ 0.85 > 0.75 and thus represents a significant separationbetween classical and quantum information theory.
Mark M. Wilde (LSU) 54 / 113
![Page 215: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/215.jpg)
Quantum strategy
Allow Alice and Bob to share two qubits in the state |Φ〉〈Φ|AB beforethe game starts.
If Alice receives x = 0, then she performs a measurement of Z . If shereceives x = 1, then she performs a measurement of X . In each case,she reports the outcome as a.
If Bob receives y = 0, then he performs a measurement of(X + Z )/
√2. If he receives y = 1, then he performs a measurement
of (Z − X )/√
2. In each case, he reports the outcome as b.
This quantum strategy has a winning probability ofcos2(π/8) ≈ 0.85 > 0.75 and thus represents a significant separationbetween classical and quantum information theory.
Mark M. Wilde (LSU) 54 / 113
![Page 216: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/216.jpg)
Quantum strategy
Allow Alice and Bob to share two qubits in the state |Φ〉〈Φ|AB beforethe game starts.
If Alice receives x = 0, then she performs a measurement of Z . If shereceives x = 1, then she performs a measurement of X . In each case,she reports the outcome as a.
If Bob receives y = 0, then he performs a measurement of(X + Z )/
√2. If he receives y = 1, then he performs a measurement
of (Z − X )/√
2. In each case, he reports the outcome as b.
This quantum strategy has a winning probability ofcos2(π/8) ≈ 0.85 > 0.75 and thus represents a significant separationbetween classical and quantum information theory.
Mark M. Wilde (LSU) 54 / 113
![Page 217: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/217.jpg)
Loophole-free Bell test...
Picture of loophole-free Bell test at TU Delft(Image taken from http://hansonlab.tudelft.nl/loophole-free-bell-test/)
Mark M. Wilde (LSU) 55 / 113
![Page 218: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/218.jpg)
Three fundamental protocols
The three important noiseless protocols in quantum informationtheory are entanglement distribution, super-dense coding, andquantum teleportation.
They are the building blocks for later core quantum communicationprotocols, in which we replace a noiseless resource with a noisy one.
Mark M. Wilde (LSU) 56 / 113
![Page 219: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/219.jpg)
Three fundamental protocols
The three important noiseless protocols in quantum informationtheory are entanglement distribution, super-dense coding, andquantum teleportation.
They are the building blocks for later core quantum communicationprotocols, in which we replace a noiseless resource with a noisy one.
Mark M. Wilde (LSU) 56 / 113
![Page 220: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/220.jpg)
Three fundamental protocols
The three important noiseless protocols in quantum informationtheory are entanglement distribution, super-dense coding, andquantum teleportation.
They are the building blocks for later core quantum communicationprotocols, in which we replace a noiseless resource with a noisy one.
Mark M. Wilde (LSU) 56 / 113
![Page 221: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/221.jpg)
Communication resources
Resources
Let [c → c] denote a noiseless classical bit channel from Alice(sender) to Bob (receiver), which performs the following mapping ona qubit density matrix:
ρ =
[ρ00 ρ01
ρ10 ρ11
]→ 1
2ρ+
1
2ZρZ =
[ρ00 00 ρ11
].
Let [q → q] denote a noiseless quantum bit channel from Alice toBob, which perfectly preserves a qubit density matrix.
Let [qq] denote a noiseless ebit shared between Alice and Bob, whichis a maximally entangled state |Φ〉〈Φ|AB .
Entanglement distribution, super-dense coding, and teleportation arenon-trivial protocols for combining these resources.
Mark M. Wilde (LSU) 57 / 113
![Page 222: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/222.jpg)
Communication resources
Resources
Let [c → c] denote a noiseless classical bit channel from Alice(sender) to Bob (receiver), which performs the following mapping ona qubit density matrix:
ρ =
[ρ00 ρ01
ρ10 ρ11
]→ 1
2ρ+
1
2ZρZ =
[ρ00 00 ρ11
].
Let [q → q] denote a noiseless quantum bit channel from Alice toBob, which perfectly preserves a qubit density matrix.
Let [qq] denote a noiseless ebit shared between Alice and Bob, whichis a maximally entangled state |Φ〉〈Φ|AB .
Entanglement distribution, super-dense coding, and teleportation arenon-trivial protocols for combining these resources.
Mark M. Wilde (LSU) 57 / 113
![Page 223: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/223.jpg)
Communication resources
Resources
Let [c → c] denote a noiseless classical bit channel from Alice(sender) to Bob (receiver), which performs the following mapping ona qubit density matrix:
ρ =
[ρ00 ρ01
ρ10 ρ11
]→ 1
2ρ+
1
2ZρZ =
[ρ00 00 ρ11
].
Let [q → q] denote a noiseless quantum bit channel from Alice toBob, which perfectly preserves a qubit density matrix.
Let [qq] denote a noiseless ebit shared between Alice and Bob, whichis a maximally entangled state |Φ〉〈Φ|AB .
Entanglement distribution, super-dense coding, and teleportation arenon-trivial protocols for combining these resources.
Mark M. Wilde (LSU) 57 / 113
![Page 224: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/224.jpg)
Communication resources
Resources
Let [c → c] denote a noiseless classical bit channel from Alice(sender) to Bob (receiver), which performs the following mapping ona qubit density matrix:
ρ =
[ρ00 ρ01
ρ10 ρ11
]→ 1
2ρ+
1
2ZρZ =
[ρ00 00 ρ11
].
Let [q → q] denote a noiseless quantum bit channel from Alice toBob, which perfectly preserves a qubit density matrix.
Let [qq] denote a noiseless ebit shared between Alice and Bob, whichis a maximally entangled state |Φ〉〈Φ|AB .
Entanglement distribution, super-dense coding, and teleportation arenon-trivial protocols for combining these resources.
Mark M. Wilde (LSU) 57 / 113
![Page 225: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/225.jpg)
Communication resources
Resources
Let [c → c] denote a noiseless classical bit channel from Alice(sender) to Bob (receiver), which performs the following mapping ona qubit density matrix:
ρ =
[ρ00 ρ01
ρ10 ρ11
]→ 1
2ρ+
1
2ZρZ =
[ρ00 00 ρ11
].
Let [q → q] denote a noiseless quantum bit channel from Alice toBob, which perfectly preserves a qubit density matrix.
Let [qq] denote a noiseless ebit shared between Alice and Bob, whichis a maximally entangled state |Φ〉〈Φ|AB .
Entanglement distribution, super-dense coding, and teleportation arenon-trivial protocols for combining these resources.
Mark M. Wilde (LSU) 57 / 113
![Page 226: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/226.jpg)
Communication resources
Resources
Let [c → c] denote a noiseless classical bit channel from Alice(sender) to Bob (receiver), which performs the following mapping ona qubit density matrix:
ρ =
[ρ00 ρ01
ρ10 ρ11
]→ 1
2ρ+
1
2ZρZ =
[ρ00 00 ρ11
].
Let [q → q] denote a noiseless quantum bit channel from Alice toBob, which perfectly preserves a qubit density matrix.
Let [qq] denote a noiseless ebit shared between Alice and Bob, whichis a maximally entangled state |Φ〉〈Φ|AB .
Entanglement distribution, super-dense coding, and teleportation arenon-trivial protocols for combining these resources.
Mark M. Wilde (LSU) 57 / 113
![Page 227: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/227.jpg)
Preparing a maximally entangled state of two qubits
How to prepare a maximally entangled state?
Alice begins by preparing two qubits in the tensor-product state:
|0〉〈0|A ⊗ |0〉〈0|A′ .
Let H = 1√2
[1 11 −1
], which is a unitary matrix. Alice performs the
unitary channel H(·)H† on her system A, leading to the global state
HA|0〉〈0|AH†A ⊗ |0〉〈0|A′ .
Alice performs CNOT = |0〉〈0|A ⊗ IA′ + |1〉〈1|A ⊗ XA′ . This is aunitary called controlled-NOT, because it flips the second bit if andonly if the first bit is one (these actions are done in superposition).
After doing this, the state on AA′ becomes |Φ〉〈Φ|AA′ .
Mark M. Wilde (LSU) 58 / 113
![Page 228: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/228.jpg)
Preparing a maximally entangled state of two qubits
How to prepare a maximally entangled state?
Alice begins by preparing two qubits in the tensor-product state:
|0〉〈0|A ⊗ |0〉〈0|A′ .
Let H = 1√2
[1 11 −1
], which is a unitary matrix. Alice performs the
unitary channel H(·)H† on her system A, leading to the global state
HA|0〉〈0|AH†A ⊗ |0〉〈0|A′ .
Alice performs CNOT = |0〉〈0|A ⊗ IA′ + |1〉〈1|A ⊗ XA′ . This is aunitary called controlled-NOT, because it flips the second bit if andonly if the first bit is one (these actions are done in superposition).
After doing this, the state on AA′ becomes |Φ〉〈Φ|AA′ .
Mark M. Wilde (LSU) 58 / 113
![Page 229: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/229.jpg)
Preparing a maximally entangled state of two qubits
How to prepare a maximally entangled state?
Alice begins by preparing two qubits in the tensor-product state:
|0〉〈0|A ⊗ |0〉〈0|A′ .
Let H = 1√2
[1 11 −1
], which is a unitary matrix. Alice performs the
unitary channel H(·)H† on her system A, leading to the global state
HA|0〉〈0|AH†A ⊗ |0〉〈0|A′ .
Alice performs CNOT = |0〉〈0|A ⊗ IA′ + |1〉〈1|A ⊗ XA′ . This is aunitary called controlled-NOT, because it flips the second bit if andonly if the first bit is one (these actions are done in superposition).
After doing this, the state on AA′ becomes |Φ〉〈Φ|AA′ .
Mark M. Wilde (LSU) 58 / 113
![Page 230: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/230.jpg)
Preparing a maximally entangled state of two qubits
How to prepare a maximally entangled state?
Alice begins by preparing two qubits in the tensor-product state:
|0〉〈0|A ⊗ |0〉〈0|A′ .
Let H = 1√2
[1 11 −1
], which is a unitary matrix. Alice performs the
unitary channel H(·)H† on her system A, leading to the global state
HA|0〉〈0|AH†A ⊗ |0〉〈0|A′ .
Alice performs CNOT = |0〉〈0|A ⊗ IA′ + |1〉〈1|A ⊗ XA′ . This is aunitary called controlled-NOT, because it flips the second bit if andonly if the first bit is one (these actions are done in superposition).
After doing this, the state on AA′ becomes |Φ〉〈Φ|AA′ .
Mark M. Wilde (LSU) 58 / 113
![Page 231: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/231.jpg)
Preparing a maximally entangled state of two qubits
How to prepare a maximally entangled state?
Alice begins by preparing two qubits in the tensor-product state:
|0〉〈0|A ⊗ |0〉〈0|A′ .
Let H = 1√2
[1 11 −1
], which is a unitary matrix. Alice performs the
unitary channel H(·)H† on her system A, leading to the global state
HA|0〉〈0|AH†A ⊗ |0〉〈0|A′ .
Alice performs CNOT = |0〉〈0|A ⊗ IA′ + |1〉〈1|A ⊗ XA′ . This is aunitary called controlled-NOT, because it flips the second bit if andonly if the first bit is one (these actions are done in superposition).
After doing this, the state on AA′ becomes |Φ〉〈Φ|AA′ .
Mark M. Wilde (LSU) 58 / 113
![Page 232: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/232.jpg)
Preparing a maximally entangled state of two qubits
How to prepare a maximally entangled state?
Alice begins by preparing two qubits in the tensor-product state:
|0〉〈0|A ⊗ |0〉〈0|A′ .
Let H = 1√2
[1 11 −1
], which is a unitary matrix. Alice performs the
unitary channel H(·)H† on her system A, leading to the global state
HA|0〉〈0|AH†A ⊗ |0〉〈0|A′ .
Alice performs CNOT = |0〉〈0|A ⊗ IA′ + |1〉〈1|A ⊗ XA′ . This is aunitary called controlled-NOT, because it flips the second bit if andonly if the first bit is one (these actions are done in superposition).
After doing this, the state on AA′ becomes |Φ〉〈Φ|AA′ .
Mark M. Wilde (LSU) 58 / 113
![Page 233: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/233.jpg)
Entanglement distribution
A’
B
id
|0〉
|0〉
A
A’→B
H
Alice performs local operations (the Hadamard and CNOT) andconsumes one use of a noiseless qubit channel to generate onenoiseless ebit |Φ〉〈Φ|AB shared with Bob.
Resource inequality: [q → q] ≥ [qq].
Mark M. Wilde (LSU) 59 / 113
![Page 234: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/234.jpg)
Entanglement distribution
A’
B
id
|0〉
|0〉
A
A’→B
H
Alice performs local operations (the Hadamard and CNOT) andconsumes one use of a noiseless qubit channel to generate onenoiseless ebit |Φ〉〈Φ|AB shared with Bob.
Resource inequality: [q → q] ≥ [qq].
Mark M. Wilde (LSU) 59 / 113
![Page 235: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/235.jpg)
Entanglement distribution
A’
B
id
|0〉
|0〉
A
A’→B
H
Alice performs local operations (the Hadamard and CNOT) andconsumes one use of a noiseless qubit channel to generate onenoiseless ebit |Φ〉〈Φ|AB shared with Bob.
Resource inequality: [q → q] ≥ [qq].
Mark M. Wilde (LSU) 59 / 113
![Page 236: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/236.jpg)
Bell states
Consider that, for a 2× 2 matrix MB ,
〈Φ|AB IA ⊗MB |Φ〉AB =1
2Tr{MB}.
I has trace 2 and Pauli matrices X , Y , and Z are traceless.Multiplying any two of them of them gives another Pauli matrix.
These facts imply that the following set forms an orthonormal basis:
{|Φ〉AB , XA|Φ〉AB , ZA|Φ〉AB , ZAXA|Φ〉AB}.
So the following states are perfectly distinguishable:
{|Φ〉〈Φ|AB , XA|Φ〉〈Φ|ABXA, ZA|Φ〉〈Φ|ABZA, ZAXA|Φ〉〈Φ|ABXAZA}.
Mark M. Wilde (LSU) 60 / 113
![Page 237: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/237.jpg)
Bell states
Consider that, for a 2× 2 matrix MB ,
〈Φ|AB IA ⊗MB |Φ〉AB =1
2Tr{MB}.
I has trace 2 and Pauli matrices X , Y , and Z are traceless.Multiplying any two of them of them gives another Pauli matrix.
These facts imply that the following set forms an orthonormal basis:
{|Φ〉AB , XA|Φ〉AB , ZA|Φ〉AB , ZAXA|Φ〉AB}.
So the following states are perfectly distinguishable:
{|Φ〉〈Φ|AB , XA|Φ〉〈Φ|ABXA, ZA|Φ〉〈Φ|ABZA, ZAXA|Φ〉〈Φ|ABXAZA}.
Mark M. Wilde (LSU) 60 / 113
![Page 238: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/238.jpg)
Bell states
Consider that, for a 2× 2 matrix MB ,
〈Φ|AB IA ⊗MB |Φ〉AB =1
2Tr{MB}.
I has trace 2 and Pauli matrices X , Y , and Z are traceless.Multiplying any two of them of them gives another Pauli matrix.
These facts imply that the following set forms an orthonormal basis:
{|Φ〉AB , XA|Φ〉AB , ZA|Φ〉AB , ZAXA|Φ〉AB}.
So the following states are perfectly distinguishable:
{|Φ〉〈Φ|AB , XA|Φ〉〈Φ|ABXA, ZA|Φ〉〈Φ|ABZA, ZAXA|Φ〉〈Φ|ABXAZA}.
Mark M. Wilde (LSU) 60 / 113
![Page 239: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/239.jpg)
Bell states
Consider that, for a 2× 2 matrix MB ,
〈Φ|AB IA ⊗MB |Φ〉AB =1
2Tr{MB}.
I has trace 2 and Pauli matrices X , Y , and Z are traceless.Multiplying any two of them of them gives another Pauli matrix.
These facts imply that the following set forms an orthonormal basis:
{|Φ〉AB , XA|Φ〉AB , ZA|Φ〉AB , ZAXA|Φ〉AB}.
So the following states are perfectly distinguishable:
{|Φ〉〈Φ|AB , XA|Φ〉〈Φ|ABXA, ZA|Φ〉〈Φ|ABZA, ZAXA|Φ〉〈Φ|ABXAZA}.
Mark M. Wilde (LSU) 60 / 113
![Page 240: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/240.jpg)
Bell states
Consider that, for a 2× 2 matrix MB ,
〈Φ|AB IA ⊗MB |Φ〉AB =1
2Tr{MB}.
I has trace 2 and Pauli matrices X , Y , and Z are traceless.Multiplying any two of them of them gives another Pauli matrix.
These facts imply that the following set forms an orthonormal basis:
{|Φ〉AB , XA|Φ〉AB , ZA|Φ〉AB , ZAXA|Φ〉AB}.
So the following states are perfectly distinguishable:
{|Φ〉〈Φ|AB , XA|Φ〉〈Φ|ABXA, ZA|Φ〉〈Φ|ABZA, ZAXA|Φ〉〈Φ|ABXAZA}.
Mark M. Wilde (LSU) 60 / 113
![Page 241: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/241.jpg)
Bell measurement
The measurement channel that distinguishes these states is called theBell measurement:
ρAB →Tr{|Φ〉〈Φ|ABρAB}|00〉〈00|+ Tr{XA|Φ〉〈Φ|ABXAρAB}|01〉〈01|+ Tr{ZA|Φ〉〈Φ|ABZAρAB}|10〉〈10|+ Tr{ZAXA|Φ〉〈Φ|ABXAZAρAB}|11〉〈11|.
This measurement can be implemented on a quantum computer byperforming controlled-NOT from A to B, Hadamard on A, and thenmeasuring A and B in the standard basis.
Mark M. Wilde (LSU) 61 / 113
![Page 242: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/242.jpg)
Bell measurement
The measurement channel that distinguishes these states is called theBell measurement:
ρAB →Tr{|Φ〉〈Φ|ABρAB}|00〉〈00|+ Tr{XA|Φ〉〈Φ|ABXAρAB}|01〉〈01|+ Tr{ZA|Φ〉〈Φ|ABZAρAB}|10〉〈10|+ Tr{ZAXA|Φ〉〈Φ|ABXAZAρAB}|11〉〈11|.
This measurement can be implemented on a quantum computer byperforming controlled-NOT from A to B, Hadamard on A, and thenmeasuring A and B in the standard basis.
Mark M. Wilde (LSU) 61 / 113
![Page 243: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/243.jpg)
Bell measurement
The measurement channel that distinguishes these states is called theBell measurement:
ρAB →Tr{|Φ〉〈Φ|ABρAB}|00〉〈00|+ Tr{XA|Φ〉〈Φ|ABXAρAB}|01〉〈01|+ Tr{ZA|Φ〉〈Φ|ABZAρAB}|10〉〈10|+ Tr{ZAXA|Φ〉〈Φ|ABXAZAρAB}|11〉〈11|.
This measurement can be implemented on a quantum computer byperforming controlled-NOT from A to B, Hadamard on A, and thenmeasuring A and B in the standard basis.
Mark M. Wilde (LSU) 61 / 113
![Page 244: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/244.jpg)
Super-dense coding
X Z
Bell Measurement
Conditional Operations
QubitChannel
x1x2
x1x2
|Ф 〉+AB
Alice and Bob share an ebit. Alice would like to transmit two classicalbits x1x2 to Bob. She performs a Pauli rotation conditioned on x1x2
and sends her share of the ebit over a noiseless qubit channel. Bobthen performs a Bell measurement to get x1x2.
Resource inequality: [q → q] + [qq] ≥ 2[c → c].
Mark M. Wilde (LSU) 62 / 113
![Page 245: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/245.jpg)
Super-dense coding
X Z
Bell Measurement
Conditional Operations
QubitChannel
x1x2
x1x2
|Ф 〉+AB
Alice and Bob share an ebit. Alice would like to transmit two classicalbits x1x2 to Bob. She performs a Pauli rotation conditioned on x1x2
and sends her share of the ebit over a noiseless qubit channel. Bobthen performs a Bell measurement to get x1x2.
Resource inequality: [q → q] + [qq] ≥ 2[c → c].
Mark M. Wilde (LSU) 62 / 113
![Page 246: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/246.jpg)
Super-dense coding
X Z
Bell Measurement
Conditional Operations
QubitChannel
x1x2
x1x2
|Ф 〉+AB
Alice and Bob share an ebit. Alice would like to transmit two classicalbits x1x2 to Bob. She performs a Pauli rotation conditioned on x1x2
and sends her share of the ebit over a noiseless qubit channel. Bobthen performs a Bell measurement to get x1x2.
Resource inequality: [q → q] + [qq] ≥ 2[c → c].
Mark M. Wilde (LSU) 62 / 113
![Page 247: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/247.jpg)
Algebraic trick for quantum teleportation
Let |ψ〉〈ψ| be the state of a qubit where |ψ〉 = α|0〉+ β|1〉.By using the algebra of the tensor product, can show that
|ψ〉A′ |Φ〉AB ∝ |Φ〉A′A |ψ〉B + XA |Φ〉A′A XB |ψ〉B+ ZA |Φ〉A′A ZB |ψ〉B + ZAXA |Φ〉A′A XBZB |ψ〉B .
Performing the Bell measurement channel on systems AA′ leads tothe following state:
1
4
[|00〉〈00|AA′ ⊗ |ψ〉〈ψ|B + |01〉〈01|AA′ ⊗ XB |ψ〉〈ψ|BXB
+ |10〉〈10|AA′ ⊗ ZB |ψ〉〈ψ|BZB
+ |11〉〈11|AA′ ⊗ XBZB |ψ〉〈ψ|BZBXB
].
Alice then sends the two classical bits in AA′ to Bob. Bob can thenundo the Pauli rotations and recover the state |ψ〉〈ψ|B .
Mark M. Wilde (LSU) 63 / 113
![Page 248: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/248.jpg)
Algebraic trick for quantum teleportation
Let |ψ〉〈ψ| be the state of a qubit where |ψ〉 = α|0〉+ β|1〉.
By using the algebra of the tensor product, can show that
|ψ〉A′ |Φ〉AB ∝ |Φ〉A′A |ψ〉B + XA |Φ〉A′A XB |ψ〉B+ ZA |Φ〉A′A ZB |ψ〉B + ZAXA |Φ〉A′A XBZB |ψ〉B .
Performing the Bell measurement channel on systems AA′ leads tothe following state:
1
4
[|00〉〈00|AA′ ⊗ |ψ〉〈ψ|B + |01〉〈01|AA′ ⊗ XB |ψ〉〈ψ|BXB
+ |10〉〈10|AA′ ⊗ ZB |ψ〉〈ψ|BZB
+ |11〉〈11|AA′ ⊗ XBZB |ψ〉〈ψ|BZBXB
].
Alice then sends the two classical bits in AA′ to Bob. Bob can thenundo the Pauli rotations and recover the state |ψ〉〈ψ|B .
Mark M. Wilde (LSU) 63 / 113
![Page 249: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/249.jpg)
Algebraic trick for quantum teleportation
Let |ψ〉〈ψ| be the state of a qubit where |ψ〉 = α|0〉+ β|1〉.By using the algebra of the tensor product, can show that
|ψ〉A′ |Φ〉AB ∝ |Φ〉A′A |ψ〉B + XA |Φ〉A′A XB |ψ〉B+ ZA |Φ〉A′A ZB |ψ〉B + ZAXA |Φ〉A′A XBZB |ψ〉B .
Performing the Bell measurement channel on systems AA′ leads tothe following state:
1
4
[|00〉〈00|AA′ ⊗ |ψ〉〈ψ|B + |01〉〈01|AA′ ⊗ XB |ψ〉〈ψ|BXB
+ |10〉〈10|AA′ ⊗ ZB |ψ〉〈ψ|BZB
+ |11〉〈11|AA′ ⊗ XBZB |ψ〉〈ψ|BZBXB
].
Alice then sends the two classical bits in AA′ to Bob. Bob can thenundo the Pauli rotations and recover the state |ψ〉〈ψ|B .
Mark M. Wilde (LSU) 63 / 113
![Page 250: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/250.jpg)
Algebraic trick for quantum teleportation
Let |ψ〉〈ψ| be the state of a qubit where |ψ〉 = α|0〉+ β|1〉.By using the algebra of the tensor product, can show that
|ψ〉A′ |Φ〉AB ∝ |Φ〉A′A |ψ〉B + XA |Φ〉A′A XB |ψ〉B+ ZA |Φ〉A′A ZB |ψ〉B + ZAXA |Φ〉A′A XBZB |ψ〉B .
Performing the Bell measurement channel on systems AA′ leads tothe following state:
1
4
[|00〉〈00|AA′ ⊗ |ψ〉〈ψ|B + |01〉〈01|AA′ ⊗ XB |ψ〉〈ψ|BXB
+ |10〉〈10|AA′ ⊗ ZB |ψ〉〈ψ|BZB
+ |11〉〈11|AA′ ⊗ XBZB |ψ〉〈ψ|BZBXB
].
Alice then sends the two classical bits in AA′ to Bob. Bob can thenundo the Pauli rotations and recover the state |ψ〉〈ψ|B .
Mark M. Wilde (LSU) 63 / 113
![Page 251: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/251.jpg)
Algebraic trick for quantum teleportation
Let |ψ〉〈ψ| be the state of a qubit where |ψ〉 = α|0〉+ β|1〉.By using the algebra of the tensor product, can show that
|ψ〉A′ |Φ〉AB ∝ |Φ〉A′A |ψ〉B + XA |Φ〉A′A XB |ψ〉B+ ZA |Φ〉A′A ZB |ψ〉B + ZAXA |Φ〉A′A XBZB |ψ〉B .
Performing the Bell measurement channel on systems AA′ leads tothe following state:
1
4
[|00〉〈00|AA′ ⊗ |ψ〉〈ψ|B + |01〉〈01|AA′ ⊗ XB |ψ〉〈ψ|BXB
+ |10〉〈10|AA′ ⊗ ZB |ψ〉〈ψ|BZB
+ |11〉〈11|AA′ ⊗ XBZB |ψ〉〈ψ|BZBXB
].
Alice then sends the two classical bits in AA′ to Bob. Bob can thenundo the Pauli rotations and recover the state |ψ〉〈ψ|B .
Mark M. Wilde (LSU) 63 / 113
![Page 252: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/252.jpg)
Teleportation
X Z
Bell Measurement
Conditional Operations
Two ClassicalChannels
|Ф 〉+AB
A’|ψ〉
B|ψ〉
Alice would like to transmit an arbitrary quantum state |ψ〉〈ψ|A′ toBob. Alice and Bob share an ebit before the protocol begins. Alicecan “teleport” her quantum state to Bob by consuming theentanglement and two uses of a noiseless classical bit channel.
Resource inequality: 2[c → c] + [qq] ≥ [q → q].
Mark M. Wilde (LSU) 64 / 113
![Page 253: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/253.jpg)
Teleportation
X Z
Bell Measurement
Conditional Operations
Two ClassicalChannels
|Ф 〉+AB
A’|ψ〉
B|ψ〉
Alice would like to transmit an arbitrary quantum state |ψ〉〈ψ|A′ toBob. Alice and Bob share an ebit before the protocol begins. Alicecan “teleport” her quantum state to Bob by consuming theentanglement and two uses of a noiseless classical bit channel.
Resource inequality: 2[c → c] + [qq] ≥ [q → q].
Mark M. Wilde (LSU) 64 / 113
![Page 254: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/254.jpg)
Teleportation
X Z
Bell Measurement
Conditional Operations
Two ClassicalChannels
|Ф 〉+AB
A’|ψ〉
B|ψ〉
Alice would like to transmit an arbitrary quantum state |ψ〉〈ψ|A′ toBob. Alice and Bob share an ebit before the protocol begins. Alicecan “teleport” her quantum state to Bob by consuming theentanglement and two uses of a noiseless classical bit channel.
Resource inequality: 2[c → c] + [qq] ≥ [q → q].
Mark M. Wilde (LSU) 64 / 113
![Page 255: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/255.jpg)
Teleportation between Canary Islands...
Teleportation between two Canary Islands 143 km apart. Green lasers wereused only for stabilization—invisible infrared photons were teleported(Image taken from http://www.ing.iac.es/PR/press/quantum.html)
Mark M. Wilde (LSU) 65 / 113
![Page 256: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/256.jpg)
Distance measures
Mark M. Wilde (LSU) 66 / 113
![Page 257: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/257.jpg)
Function of a diagonalizable matrix
If an n × n matrix D is diagonal with entries d1, . . . , dn, then for afunction f , we define
f (D) =
g(d1) 0 · · · 0
0 g(d2)...
.... . . 0
0 · · · 0 g(dn)
where g(x) = f (x) if x 6= 0 and g(x) = 0 otherwise.
If a matrix A is diagonalizable as A = KDK−1, then for a function f ,we define
f (A) = Kf (D)K−1.
Evaluating the function only on the support of the matrix allows forfunctions such as f (x) = x−1 and f (x) = log x .
Mark M. Wilde (LSU) 67 / 113
![Page 258: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/258.jpg)
Function of a diagonalizable matrix
If an n × n matrix D is diagonal with entries d1, . . . , dn, then for afunction f , we define
f (D) =
g(d1) 0 · · · 0
0 g(d2)...
.... . . 0
0 · · · 0 g(dn)
where g(x) = f (x) if x 6= 0 and g(x) = 0 otherwise.
If a matrix A is diagonalizable as A = KDK−1, then for a function f ,we define
f (A) = Kf (D)K−1.
Evaluating the function only on the support of the matrix allows forfunctions such as f (x) = x−1 and f (x) = log x .
Mark M. Wilde (LSU) 67 / 113
![Page 259: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/259.jpg)
Function of a diagonalizable matrix
If an n × n matrix D is diagonal with entries d1, . . . , dn, then for afunction f , we define
f (D) =
g(d1) 0 · · · 0
0 g(d2)...
.... . . 0
0 · · · 0 g(dn)
where g(x) = f (x) if x 6= 0 and g(x) = 0 otherwise.
If a matrix A is diagonalizable as A = KDK−1, then for a function f ,we define
f (A) = Kf (D)K−1.
Evaluating the function only on the support of the matrix allows forfunctions such as f (x) = x−1 and f (x) = log x .
Mark M. Wilde (LSU) 67 / 113
![Page 260: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/260.jpg)
Function of a diagonalizable matrix
If an n × n matrix D is diagonal with entries d1, . . . , dn, then for afunction f , we define
f (D) =
g(d1) 0 · · · 0
0 g(d2)...
.... . . 0
0 · · · 0 g(dn)
where g(x) = f (x) if x 6= 0 and g(x) = 0 otherwise.
If a matrix A is diagonalizable as A = KDK−1, then for a function f ,we define
f (A) = Kf (D)K−1.
Evaluating the function only on the support of the matrix allows forfunctions such as f (x) = x−1 and f (x) = log x .
Mark M. Wilde (LSU) 67 / 113
![Page 261: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/261.jpg)
Trace distance
Define the trace norm of a matrix X by ‖X‖1 ≡ Tr{√X †X}.
Trace norm induces trace distance between two matrices X and Y :
‖X − Y ‖1.
For two density matrices ρ and σ, the following bounds hold
0 ≤ ‖ρ− σ‖1 ≤ 2.
LHS saturated iff ρ = σ and RHS iff ρ is orthogonal to σ.
For commuting ρ and σ, trace distance reduces to variational distancebetween probability distributions along diagonals.
Has an operational meaning as the bias of the optimal successprobability in a hypothesis test to distinguish ρ from σ.
Does not increase under the action of a quantum channel:
‖ρ− σ‖1 ≥ ‖N (ρ)−N (σ)‖1.
Mark M. Wilde (LSU) 68 / 113
![Page 262: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/262.jpg)
Trace distance
Define the trace norm of a matrix X by ‖X‖1 ≡ Tr{√X †X}.
Trace norm induces trace distance between two matrices X and Y :
‖X − Y ‖1.
For two density matrices ρ and σ, the following bounds hold
0 ≤ ‖ρ− σ‖1 ≤ 2.
LHS saturated iff ρ = σ and RHS iff ρ is orthogonal to σ.
For commuting ρ and σ, trace distance reduces to variational distancebetween probability distributions along diagonals.
Has an operational meaning as the bias of the optimal successprobability in a hypothesis test to distinguish ρ from σ.
Does not increase under the action of a quantum channel:
‖ρ− σ‖1 ≥ ‖N (ρ)−N (σ)‖1.
Mark M. Wilde (LSU) 68 / 113
![Page 263: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/263.jpg)
Trace distance
Define the trace norm of a matrix X by ‖X‖1 ≡ Tr{√X †X}.
Trace norm induces trace distance between two matrices X and Y :
‖X − Y ‖1.
For two density matrices ρ and σ, the following bounds hold
0 ≤ ‖ρ− σ‖1 ≤ 2.
LHS saturated iff ρ = σ and RHS iff ρ is orthogonal to σ.
For commuting ρ and σ, trace distance reduces to variational distancebetween probability distributions along diagonals.
Has an operational meaning as the bias of the optimal successprobability in a hypothesis test to distinguish ρ from σ.
Does not increase under the action of a quantum channel:
‖ρ− σ‖1 ≥ ‖N (ρ)−N (σ)‖1.
Mark M. Wilde (LSU) 68 / 113
![Page 264: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/264.jpg)
Trace distance
Define the trace norm of a matrix X by ‖X‖1 ≡ Tr{√X †X}.
Trace norm induces trace distance between two matrices X and Y :
‖X − Y ‖1.
For two density matrices ρ and σ, the following bounds hold
0 ≤ ‖ρ− σ‖1 ≤ 2.
LHS saturated iff ρ = σ and RHS iff ρ is orthogonal to σ.
For commuting ρ and σ, trace distance reduces to variational distancebetween probability distributions along diagonals.
Has an operational meaning as the bias of the optimal successprobability in a hypothesis test to distinguish ρ from σ.
Does not increase under the action of a quantum channel:
‖ρ− σ‖1 ≥ ‖N (ρ)−N (σ)‖1.
Mark M. Wilde (LSU) 68 / 113
![Page 265: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/265.jpg)
Trace distance
Define the trace norm of a matrix X by ‖X‖1 ≡ Tr{√X †X}.
Trace norm induces trace distance between two matrices X and Y :
‖X − Y ‖1.
For two density matrices ρ and σ, the following bounds hold
0 ≤ ‖ρ− σ‖1 ≤ 2.
LHS saturated iff ρ = σ and RHS iff ρ is orthogonal to σ.
For commuting ρ and σ, trace distance reduces to variational distancebetween probability distributions along diagonals.
Has an operational meaning as the bias of the optimal successprobability in a hypothesis test to distinguish ρ from σ.
Does not increase under the action of a quantum channel:
‖ρ− σ‖1 ≥ ‖N (ρ)−N (σ)‖1.
Mark M. Wilde (LSU) 68 / 113
![Page 266: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/266.jpg)
Trace distance
Define the trace norm of a matrix X by ‖X‖1 ≡ Tr{√X †X}.
Trace norm induces trace distance between two matrices X and Y :
‖X − Y ‖1.
For two density matrices ρ and σ, the following bounds hold
0 ≤ ‖ρ− σ‖1 ≤ 2.
LHS saturated iff ρ = σ and RHS iff ρ is orthogonal to σ.
For commuting ρ and σ, trace distance reduces to variational distancebetween probability distributions along diagonals.
Has an operational meaning as the bias of the optimal successprobability in a hypothesis test to distinguish ρ from σ.
Does not increase under the action of a quantum channel:
‖ρ− σ‖1 ≥ ‖N (ρ)−N (σ)‖1.
Mark M. Wilde (LSU) 68 / 113
![Page 267: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/267.jpg)
Trace distance
Define the trace norm of a matrix X by ‖X‖1 ≡ Tr{√X †X}.
Trace norm induces trace distance between two matrices X and Y :
‖X − Y ‖1.
For two density matrices ρ and σ, the following bounds hold
0 ≤ ‖ρ− σ‖1 ≤ 2.
LHS saturated iff ρ = σ and RHS iff ρ is orthogonal to σ.
For commuting ρ and σ, trace distance reduces to variational distancebetween probability distributions along diagonals.
Has an operational meaning as the bias of the optimal successprobability in a hypothesis test to distinguish ρ from σ.
Does not increase under the action of a quantum channel:
‖ρ− σ‖1 ≥ ‖N (ρ)−N (σ)‖1.
Mark M. Wilde (LSU) 68 / 113
![Page 268: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/268.jpg)
Fidelity
Fidelity F (ρ, σ) between density matrices ρ and σ is
F (ρ, σ) ≡ ‖√ρ√σ‖2
1.
For pure states |ψ〉〈ψ| and |φ〉〈φ|, reduces to squared overlap:
F (|ψ〉〈ψ|, |φ〉〈φ|) = |〈ψ|φ〉|2.
For commuting ρ and σ, reduces to Bhattacharyya coefficient ofprobability distributions along diagonals.
For density matrices ρ and σ, the following bounds hold:
0 ≤ F (ρ, σ) ≤ 1.
LHS saturated iff ρ and σ are orthogonal and RHS iff ρ = σ.
Fidelity does not decrease under the action of a quantum channel N :
F (ρ, σ) ≤ F (N (ρ),N (σ)).
Mark M. Wilde (LSU) 69 / 113
![Page 269: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/269.jpg)
Fidelity
Fidelity F (ρ, σ) between density matrices ρ and σ is
F (ρ, σ) ≡ ‖√ρ√σ‖2
1.
For pure states |ψ〉〈ψ| and |φ〉〈φ|, reduces to squared overlap:
F (|ψ〉〈ψ|, |φ〉〈φ|) = |〈ψ|φ〉|2.
For commuting ρ and σ, reduces to Bhattacharyya coefficient ofprobability distributions along diagonals.
For density matrices ρ and σ, the following bounds hold:
0 ≤ F (ρ, σ) ≤ 1.
LHS saturated iff ρ and σ are orthogonal and RHS iff ρ = σ.
Fidelity does not decrease under the action of a quantum channel N :
F (ρ, σ) ≤ F (N (ρ),N (σ)).
Mark M. Wilde (LSU) 69 / 113
![Page 270: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/270.jpg)
Fidelity
Fidelity F (ρ, σ) between density matrices ρ and σ is
F (ρ, σ) ≡ ‖√ρ√σ‖2
1.
For pure states |ψ〉〈ψ| and |φ〉〈φ|, reduces to squared overlap:
F (|ψ〉〈ψ|, |φ〉〈φ|) = |〈ψ|φ〉|2.
For commuting ρ and σ, reduces to Bhattacharyya coefficient ofprobability distributions along diagonals.
For density matrices ρ and σ, the following bounds hold:
0 ≤ F (ρ, σ) ≤ 1.
LHS saturated iff ρ and σ are orthogonal and RHS iff ρ = σ.
Fidelity does not decrease under the action of a quantum channel N :
F (ρ, σ) ≤ F (N (ρ),N (σ)).
Mark M. Wilde (LSU) 69 / 113
![Page 271: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/271.jpg)
Fidelity
Fidelity F (ρ, σ) between density matrices ρ and σ is
F (ρ, σ) ≡ ‖√ρ√σ‖2
1.
For pure states |ψ〉〈ψ| and |φ〉〈φ|, reduces to squared overlap:
F (|ψ〉〈ψ|, |φ〉〈φ|) = |〈ψ|φ〉|2.
For commuting ρ and σ, reduces to Bhattacharyya coefficient ofprobability distributions along diagonals.
For density matrices ρ and σ, the following bounds hold:
0 ≤ F (ρ, σ) ≤ 1.
LHS saturated iff ρ and σ are orthogonal and RHS iff ρ = σ.
Fidelity does not decrease under the action of a quantum channel N :
F (ρ, σ) ≤ F (N (ρ),N (σ)).
Mark M. Wilde (LSU) 69 / 113
![Page 272: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/272.jpg)
Fidelity
Fidelity F (ρ, σ) between density matrices ρ and σ is
F (ρ, σ) ≡ ‖√ρ√σ‖2
1.
For pure states |ψ〉〈ψ| and |φ〉〈φ|, reduces to squared overlap:
F (|ψ〉〈ψ|, |φ〉〈φ|) = |〈ψ|φ〉|2.
For commuting ρ and σ, reduces to Bhattacharyya coefficient ofprobability distributions along diagonals.
For density matrices ρ and σ, the following bounds hold:
0 ≤ F (ρ, σ) ≤ 1.
LHS saturated iff ρ and σ are orthogonal and RHS iff ρ = σ.
Fidelity does not decrease under the action of a quantum channel N :
F (ρ, σ) ≤ F (N (ρ),N (σ)).
Mark M. Wilde (LSU) 69 / 113
![Page 273: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/273.jpg)
Fidelity
Fidelity F (ρ, σ) between density matrices ρ and σ is
F (ρ, σ) ≡ ‖√ρ√σ‖2
1.
For pure states |ψ〉〈ψ| and |φ〉〈φ|, reduces to squared overlap:
F (|ψ〉〈ψ|, |φ〉〈φ|) = |〈ψ|φ〉|2.
For commuting ρ and σ, reduces to Bhattacharyya coefficient ofprobability distributions along diagonals.
For density matrices ρ and σ, the following bounds hold:
0 ≤ F (ρ, σ) ≤ 1.
LHS saturated iff ρ and σ are orthogonal and RHS iff ρ = σ.
Fidelity does not decrease under the action of a quantum channel N :
F (ρ, σ) ≤ F (N (ρ),N (σ)).
Mark M. Wilde (LSU) 69 / 113
![Page 274: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/274.jpg)
Uhlmann’s theorem
Uhlmann’s theorem states that
F (ρS , σS) = maxUR
|〈ψ|RSUR ⊗ IS |φ〉RS |2,
where |ψ〉RS and |φ〉RS purify ρS and σS , respectively.
A core theorem used in quantum Shannon theory, and in other areassuch as quantum complexity theory and quantum error correction.
Since it involves purifications, this theorem has no analog in classicalinformation theory.
Mark M. Wilde (LSU) 70 / 113
![Page 275: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/275.jpg)
Uhlmann’s theorem
Uhlmann’s theorem states that
F (ρS , σS) = maxUR
|〈ψ|RSUR ⊗ IS |φ〉RS |2,
where |ψ〉RS and |φ〉RS purify ρS and σS , respectively.
A core theorem used in quantum Shannon theory, and in other areassuch as quantum complexity theory and quantum error correction.
Since it involves purifications, this theorem has no analog in classicalinformation theory.
Mark M. Wilde (LSU) 70 / 113
![Page 276: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/276.jpg)
Uhlmann’s theorem
Uhlmann’s theorem states that
F (ρS , σS) = maxUR
|〈ψ|RSUR ⊗ IS |φ〉RS |2,
where |ψ〉RS and |φ〉RS purify ρS and σS , respectively.
A core theorem used in quantum Shannon theory, and in other areassuch as quantum complexity theory and quantum error correction.
Since it involves purifications, this theorem has no analog in classicalinformation theory.
Mark M. Wilde (LSU) 70 / 113
![Page 277: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/277.jpg)
Uhlmann’s theorem
Uhlmann’s theorem states that
F (ρS , σS) = maxUR
|〈ψ|RSUR ⊗ IS |φ〉RS |2,
where |ψ〉RS and |φ〉RS purify ρS and σS , respectively.
A core theorem used in quantum Shannon theory, and in other areassuch as quantum complexity theory and quantum error correction.
Since it involves purifications, this theorem has no analog in classicalinformation theory.
Mark M. Wilde (LSU) 70 / 113
![Page 278: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/278.jpg)
Relations between fidelity and trace distance
Trace distance is useful because it obeys the triangle inequality, andfidelity is useful because we have Uhlmann’s theorem.
The following inequalities relate the two measures, which allows forgoing back and forth between them:
1−√F (ρ, σ) ≤ 1
2‖ρ− σ‖1 ≤
√1− F (ρ, σ).
A distance measure which has both properties (triangle inequality andUhlmann’s theorem) is
√1− F (ρ, σ).
Mark M. Wilde (LSU) 71 / 113
![Page 279: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/279.jpg)
Relations between fidelity and trace distance
Trace distance is useful because it obeys the triangle inequality, andfidelity is useful because we have Uhlmann’s theorem.
The following inequalities relate the two measures, which allows forgoing back and forth between them:
1−√F (ρ, σ) ≤ 1
2‖ρ− σ‖1 ≤
√1− F (ρ, σ).
A distance measure which has both properties (triangle inequality andUhlmann’s theorem) is
√1− F (ρ, σ).
Mark M. Wilde (LSU) 71 / 113
![Page 280: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/280.jpg)
Relations between fidelity and trace distance
Trace distance is useful because it obeys the triangle inequality, andfidelity is useful because we have Uhlmann’s theorem.
The following inequalities relate the two measures, which allows forgoing back and forth between them:
1−√F (ρ, σ) ≤ 1
2‖ρ− σ‖1 ≤
√1− F (ρ, σ).
A distance measure which has both properties (triangle inequality andUhlmann’s theorem) is
√1− F (ρ, σ).
Mark M. Wilde (LSU) 71 / 113
![Page 281: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/281.jpg)
Relations between fidelity and trace distance
Trace distance is useful because it obeys the triangle inequality, andfidelity is useful because we have Uhlmann’s theorem.
The following inequalities relate the two measures, which allows forgoing back and forth between them:
1−√F (ρ, σ) ≤ 1
2‖ρ− σ‖1 ≤
√1− F (ρ, σ).
A distance measure which has both properties (triangle inequality andUhlmann’s theorem) is
√1− F (ρ, σ).
Mark M. Wilde (LSU) 71 / 113
![Page 282: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/282.jpg)
Information measures
Mark M. Wilde (LSU) 72 / 113
![Page 283: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/283.jpg)
Entropy and information...
Entropy and information can be discomforting...
Mark M. Wilde (LSU) 73 / 113
![Page 284: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/284.jpg)
Quantum relative entropy
One of the most fundamental information measures is the quantumrelative entropy, defined for a state ρ and a positive semi-definitematrix σ as
D(ρ‖σ) ≡ Tr{ρ[log2 ρ− log2 σ]},
when supp(ρ) ⊆ supp(σ) and as +∞ otherwise.
It does not increase under the action of a quantum channel N :
D(ρ‖σ) ≥ D(N (ρ)‖N (σ)).
If Tr{ρ} ≥ Tr{σ}, thenD(ρ‖σ) ≥ 0,
with equality holding iff ρ = σ.
Quantum Pinsker inequality: D(ρ‖σ) ≥ 12 ln 2‖ρ− σ‖
21.
Mark M. Wilde (LSU) 74 / 113
![Page 285: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/285.jpg)
Quantum relative entropy
One of the most fundamental information measures is the quantumrelative entropy, defined for a state ρ and a positive semi-definitematrix σ as
D(ρ‖σ) ≡ Tr{ρ[log2 ρ− log2 σ]},
when supp(ρ) ⊆ supp(σ) and as +∞ otherwise.
It does not increase under the action of a quantum channel N :
D(ρ‖σ) ≥ D(N (ρ)‖N (σ)).
If Tr{ρ} ≥ Tr{σ}, thenD(ρ‖σ) ≥ 0,
with equality holding iff ρ = σ.
Quantum Pinsker inequality: D(ρ‖σ) ≥ 12 ln 2‖ρ− σ‖
21.
Mark M. Wilde (LSU) 74 / 113
![Page 286: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/286.jpg)
Quantum relative entropy
One of the most fundamental information measures is the quantumrelative entropy, defined for a state ρ and a positive semi-definitematrix σ as
D(ρ‖σ) ≡ Tr{ρ[log2 ρ− log2 σ]},
when supp(ρ) ⊆ supp(σ) and as +∞ otherwise.
It does not increase under the action of a quantum channel N :
D(ρ‖σ) ≥ D(N (ρ)‖N (σ)).
If Tr{ρ} ≥ Tr{σ}, thenD(ρ‖σ) ≥ 0,
with equality holding iff ρ = σ.
Quantum Pinsker inequality: D(ρ‖σ) ≥ 12 ln 2‖ρ− σ‖
21.
Mark M. Wilde (LSU) 74 / 113
![Page 287: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/287.jpg)
Quantum relative entropy
One of the most fundamental information measures is the quantumrelative entropy, defined for a state ρ and a positive semi-definitematrix σ as
D(ρ‖σ) ≡ Tr{ρ[log2 ρ− log2 σ]},
when supp(ρ) ⊆ supp(σ) and as +∞ otherwise.
It does not increase under the action of a quantum channel N :
D(ρ‖σ) ≥ D(N (ρ)‖N (σ)).
If Tr{ρ} ≥ Tr{σ}, thenD(ρ‖σ) ≥ 0,
with equality holding iff ρ = σ.
Quantum Pinsker inequality: D(ρ‖σ) ≥ 12 ln 2‖ρ− σ‖
21.
Mark M. Wilde (LSU) 74 / 113
![Page 288: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/288.jpg)
Quantum relative entropy
One of the most fundamental information measures is the quantumrelative entropy, defined for a state ρ and a positive semi-definitematrix σ as
D(ρ‖σ) ≡ Tr{ρ[log2 ρ− log2 σ]},
when supp(ρ) ⊆ supp(σ) and as +∞ otherwise.
It does not increase under the action of a quantum channel N :
D(ρ‖σ) ≥ D(N (ρ)‖N (σ)).
If Tr{ρ} ≥ Tr{σ}, thenD(ρ‖σ) ≥ 0,
with equality holding iff ρ = σ.
Quantum Pinsker inequality: D(ρ‖σ) ≥ 12 ln 2‖ρ− σ‖
21.
Mark M. Wilde (LSU) 74 / 113
![Page 289: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/289.jpg)
Children of quantum relative entropy
Relative entropy as “parent” entropy
Many entropies can be written in terms of relative entropy:
H(A)ρ ≡ −D(ρA‖IA) = −Tr{ρA log2 ρA} (entropy)
H(A|B)ρ ≡ −D(ρAB‖IA ⊗ ρB) (conditional entropy)
I (A;B)ρ ≡ D(ρAB‖ρA ⊗ ρB) (mutual information)
I (A〉B)ρ ≡ D(ρAB‖IA ⊗ ρB) (coherent information)
Equalities
H(A|B)ρ = H(AB)ρ − H(B)ρ
I (A〉B)ρ = −H(A|B)ρ
I (A;B)ρ = H(A)ρ + H(B)ρ − H(AB)ρ
I (A;B|C )ρ ≡ H(AC )ρ + H(BC )ρ − H(ABC )ρ − H(C )ρ
I (A;B|C )ρ = H(B|C )ρ − H(B|AC )ρ
Mark M. Wilde (LSU) 75 / 113
![Page 290: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/290.jpg)
Children of quantum relative entropy
Relative entropy as “parent” entropy
Many entropies can be written in terms of relative entropy:
H(A)ρ ≡ −D(ρA‖IA) = −Tr{ρA log2 ρA} (entropy)
H(A|B)ρ ≡ −D(ρAB‖IA ⊗ ρB) (conditional entropy)
I (A;B)ρ ≡ D(ρAB‖ρA ⊗ ρB) (mutual information)
I (A〉B)ρ ≡ D(ρAB‖IA ⊗ ρB) (coherent information)
Equalities
H(A|B)ρ = H(AB)ρ − H(B)ρ
I (A〉B)ρ = −H(A|B)ρ
I (A;B)ρ = H(A)ρ + H(B)ρ − H(AB)ρ
I (A;B|C )ρ ≡ H(AC )ρ + H(BC )ρ − H(ABC )ρ − H(C )ρ
I (A;B|C )ρ = H(B|C )ρ − H(B|AC )ρ
Mark M. Wilde (LSU) 75 / 113
![Page 291: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/291.jpg)
Children of quantum relative entropy
Relative entropy as “parent” entropy
Many entropies can be written in terms of relative entropy:
H(A)ρ ≡ −D(ρA‖IA) = −Tr{ρA log2 ρA} (entropy)
H(A|B)ρ ≡ −D(ρAB‖IA ⊗ ρB) (conditional entropy)
I (A;B)ρ ≡ D(ρAB‖ρA ⊗ ρB) (mutual information)
I (A〉B)ρ ≡ D(ρAB‖IA ⊗ ρB) (coherent information)
Equalities
H(A|B)ρ = H(AB)ρ − H(B)ρ
I (A〉B)ρ = −H(A|B)ρ
I (A;B)ρ = H(A)ρ + H(B)ρ − H(AB)ρ
I (A;B|C )ρ ≡ H(AC )ρ + H(BC )ρ − H(ABC )ρ − H(C )ρ
I (A;B|C )ρ = H(B|C )ρ − H(B|AC )ρ
Mark M. Wilde (LSU) 75 / 113
![Page 292: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/292.jpg)
Children of quantum relative entropy
Relative entropy as “parent” entropy
Many entropies can be written in terms of relative entropy:
H(A)ρ ≡ −D(ρA‖IA) = −Tr{ρA log2 ρA} (entropy)
H(A|B)ρ ≡ −D(ρAB‖IA ⊗ ρB) (conditional entropy)
I (A;B)ρ ≡ D(ρAB‖ρA ⊗ ρB) (mutual information)
I (A〉B)ρ ≡ D(ρAB‖IA ⊗ ρB) (coherent information)
Equalities
H(A|B)ρ = H(AB)ρ − H(B)ρ
I (A〉B)ρ = −H(A|B)ρ
I (A;B)ρ = H(A)ρ + H(B)ρ − H(AB)ρ
I (A;B|C )ρ ≡ H(AC )ρ + H(BC )ρ − H(ABC )ρ − H(C )ρ
I (A;B|C )ρ = H(B|C )ρ − H(B|AC )ρ
Mark M. Wilde (LSU) 75 / 113
![Page 293: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/293.jpg)
Children of quantum relative entropy
Relative entropy as “parent” entropy
Many entropies can be written in terms of relative entropy:
H(A)ρ ≡ −D(ρA‖IA) = −Tr{ρA log2 ρA} (entropy)
H(A|B)ρ ≡ −D(ρAB‖IA ⊗ ρB) (conditional entropy)
I (A;B)ρ ≡ D(ρAB‖ρA ⊗ ρB) (mutual information)
I (A〉B)ρ ≡ D(ρAB‖IA ⊗ ρB) (coherent information)
Equalities
H(A|B)ρ = H(AB)ρ − H(B)ρ
I (A〉B)ρ = −H(A|B)ρ
I (A;B)ρ = H(A)ρ + H(B)ρ − H(AB)ρ
I (A;B|C )ρ ≡ H(AC )ρ + H(BC )ρ − H(ABC )ρ − H(C )ρ
I (A;B|C )ρ = H(B|C )ρ − H(B|AC )ρ
Mark M. Wilde (LSU) 75 / 113
![Page 294: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/294.jpg)
Children of quantum relative entropy
Relative entropy as “parent” entropy
Many entropies can be written in terms of relative entropy:
H(A)ρ ≡ −D(ρA‖IA) = −Tr{ρA log2 ρA} (entropy)
H(A|B)ρ ≡ −D(ρAB‖IA ⊗ ρB) (conditional entropy)
I (A;B)ρ ≡ D(ρAB‖ρA ⊗ ρB) (mutual information)
I (A〉B)ρ ≡ D(ρAB‖IA ⊗ ρB) (coherent information)
Equalities
H(A|B)ρ = H(AB)ρ − H(B)ρ
I (A〉B)ρ = −H(A|B)ρ
I (A;B)ρ = H(A)ρ + H(B)ρ − H(AB)ρ
I (A;B|C )ρ ≡ H(AC )ρ + H(BC )ρ − H(ABC )ρ − H(C )ρ
I (A;B|C )ρ = H(B|C )ρ − H(B|AC )ρ
Mark M. Wilde (LSU) 75 / 113
![Page 295: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/295.jpg)
Children of quantum relative entropy
Relative entropy as “parent” entropy
Many entropies can be written in terms of relative entropy:
H(A)ρ ≡ −D(ρA‖IA) = −Tr{ρA log2 ρA} (entropy)
H(A|B)ρ ≡ −D(ρAB‖IA ⊗ ρB) (conditional entropy)
I (A;B)ρ ≡ D(ρAB‖ρA ⊗ ρB) (mutual information)
I (A〉B)ρ ≡ D(ρAB‖IA ⊗ ρB) (coherent information)
Equalities
H(A|B)ρ = H(AB)ρ − H(B)ρ
I (A〉B)ρ = −H(A|B)ρ
I (A;B)ρ = H(A)ρ + H(B)ρ − H(AB)ρ
I (A;B|C )ρ ≡ H(AC )ρ + H(BC )ρ − H(ABC )ρ − H(C )ρ
I (A;B|C )ρ = H(B|C )ρ − H(B|AC )ρ
Mark M. Wilde (LSU) 75 / 113
![Page 296: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/296.jpg)
Children of quantum relative entropy
Relative entropy as “parent” entropy
Many entropies can be written in terms of relative entropy:
H(A)ρ ≡ −D(ρA‖IA) = −Tr{ρA log2 ρA} (entropy)
H(A|B)ρ ≡ −D(ρAB‖IA ⊗ ρB) (conditional entropy)
I (A;B)ρ ≡ D(ρAB‖ρA ⊗ ρB) (mutual information)
I (A〉B)ρ ≡ D(ρAB‖IA ⊗ ρB) (coherent information)
Equalities
H(A|B)ρ = H(AB)ρ − H(B)ρ
I (A〉B)ρ = −H(A|B)ρ
I (A;B)ρ = H(A)ρ + H(B)ρ − H(AB)ρ
I (A;B|C )ρ ≡ H(AC )ρ + H(BC )ρ − H(ABC )ρ − H(C )ρ
I (A;B|C )ρ = H(B|C )ρ − H(B|AC )ρ
Mark M. Wilde (LSU) 75 / 113
![Page 297: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/297.jpg)
Children of quantum relative entropy
Relative entropy as “parent” entropy
Many entropies can be written in terms of relative entropy:
H(A)ρ ≡ −D(ρA‖IA) = −Tr{ρA log2 ρA} (entropy)
H(A|B)ρ ≡ −D(ρAB‖IA ⊗ ρB) (conditional entropy)
I (A;B)ρ ≡ D(ρAB‖ρA ⊗ ρB) (mutual information)
I (A〉B)ρ ≡ D(ρAB‖IA ⊗ ρB) (coherent information)
Equalities
H(A|B)ρ = H(AB)ρ − H(B)ρ
I (A〉B)ρ = −H(A|B)ρ
I (A;B)ρ = H(A)ρ + H(B)ρ − H(AB)ρ
I (A;B|C )ρ ≡ H(AC )ρ + H(BC )ρ − H(ABC )ρ − H(C )ρ
I (A;B|C )ρ = H(B|C )ρ − H(B|AC )ρ
Mark M. Wilde (LSU) 75 / 113
![Page 298: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/298.jpg)
Children of quantum relative entropy
Relative entropy as “parent” entropy
Many entropies can be written in terms of relative entropy:
H(A)ρ ≡ −D(ρA‖IA) = −Tr{ρA log2 ρA} (entropy)
H(A|B)ρ ≡ −D(ρAB‖IA ⊗ ρB) (conditional entropy)
I (A;B)ρ ≡ D(ρAB‖ρA ⊗ ρB) (mutual information)
I (A〉B)ρ ≡ D(ρAB‖IA ⊗ ρB) (coherent information)
Equalities
H(A|B)ρ = H(AB)ρ − H(B)ρ
I (A〉B)ρ = −H(A|B)ρ
I (A;B)ρ = H(A)ρ + H(B)ρ − H(AB)ρ
I (A;B|C )ρ ≡ H(AC )ρ + H(BC )ρ − H(ABC )ρ − H(C )ρ
I (A;B|C )ρ = H(B|C )ρ − H(B|AC )ρ
Mark M. Wilde (LSU) 75 / 113
![Page 299: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/299.jpg)
Children of quantum relative entropy
Relative entropy as “parent” entropy
Many entropies can be written in terms of relative entropy:
H(A)ρ ≡ −D(ρA‖IA) = −Tr{ρA log2 ρA} (entropy)
H(A|B)ρ ≡ −D(ρAB‖IA ⊗ ρB) (conditional entropy)
I (A;B)ρ ≡ D(ρAB‖ρA ⊗ ρB) (mutual information)
I (A〉B)ρ ≡ D(ρAB‖IA ⊗ ρB) (coherent information)
Equalities
H(A|B)ρ = H(AB)ρ − H(B)ρ
I (A〉B)ρ = −H(A|B)ρ
I (A;B)ρ = H(A)ρ + H(B)ρ − H(AB)ρ
I (A;B|C )ρ ≡ H(AC )ρ + H(BC )ρ − H(ABC )ρ − H(C )ρ
I (A;B|C )ρ = H(B|C )ρ − H(B|AC )ρ
Mark M. Wilde (LSU) 75 / 113
![Page 300: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/300.jpg)
Children of quantum relative entropy
Relative entropy as “parent” entropy
Many entropies can be written in terms of relative entropy:
H(A)ρ ≡ −D(ρA‖IA) = −Tr{ρA log2 ρA} (entropy)
H(A|B)ρ ≡ −D(ρAB‖IA ⊗ ρB) (conditional entropy)
I (A;B)ρ ≡ D(ρAB‖ρA ⊗ ρB) (mutual information)
I (A〉B)ρ ≡ D(ρAB‖IA ⊗ ρB) (coherent information)
Equalities
H(A|B)ρ = H(AB)ρ − H(B)ρ
I (A〉B)ρ = −H(A|B)ρ
I (A;B)ρ = H(A)ρ + H(B)ρ − H(AB)ρ
I (A;B|C )ρ ≡ H(AC )ρ + H(BC )ρ − H(ABC )ρ − H(C )ρ
I (A;B|C )ρ = H(B|C )ρ − H(B|AC )ρ
Mark M. Wilde (LSU) 75 / 113
![Page 301: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/301.jpg)
Evaluating quantum entropy
How do we evaluate the formula for quantum entropy of a state ρA?
Consider spectral decomposition:
ρA =∑x
pX (x)|x〉〈x |A.
Then, with η(x) = −x log2(x),
H(A)ρ = Tr{η(ρA)} = Tr
{∑x
η(pX (x))|x〉〈x |A
}=∑x
η(pX (x)) Tr{|x〉〈x |A} =∑x
η(pX (x)) = H(pX ).
Quantum entropy of ρA is equal to Shannon entropy of eigenvalues.
⇒ Entropy of a pure state is equal to zero.
Mark M. Wilde (LSU) 76 / 113
![Page 302: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/302.jpg)
Evaluating quantum entropy
How do we evaluate the formula for quantum entropy of a state ρA?
Consider spectral decomposition:
ρA =∑x
pX (x)|x〉〈x |A.
Then, with η(x) = −x log2(x),
H(A)ρ = Tr{η(ρA)} = Tr
{∑x
η(pX (x))|x〉〈x |A
}=∑x
η(pX (x)) Tr{|x〉〈x |A} =∑x
η(pX (x)) = H(pX ).
Quantum entropy of ρA is equal to Shannon entropy of eigenvalues.
⇒ Entropy of a pure state is equal to zero.
Mark M. Wilde (LSU) 76 / 113
![Page 303: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/303.jpg)
Evaluating quantum entropy
How do we evaluate the formula for quantum entropy of a state ρA?
Consider spectral decomposition:
ρA =∑x
pX (x)|x〉〈x |A.
Then, with η(x) = −x log2(x),
H(A)ρ = Tr{η(ρA)} = Tr
{∑x
η(pX (x))|x〉〈x |A
}=∑x
η(pX (x)) Tr{|x〉〈x |A} =∑x
η(pX (x)) = H(pX ).
Quantum entropy of ρA is equal to Shannon entropy of eigenvalues.
⇒ Entropy of a pure state is equal to zero.
Mark M. Wilde (LSU) 76 / 113
![Page 304: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/304.jpg)
Evaluating quantum entropy
How do we evaluate the formula for quantum entropy of a state ρA?
Consider spectral decomposition:
ρA =∑x
pX (x)|x〉〈x |A.
Then, with η(x) = −x log2(x),
H(A)ρ = Tr{η(ρA)} = Tr
{∑x
η(pX (x))|x〉〈x |A
}=∑x
η(pX (x)) Tr{|x〉〈x |A} =∑x
η(pX (x)) = H(pX ).
Quantum entropy of ρA is equal to Shannon entropy of eigenvalues.
⇒ Entropy of a pure state is equal to zero.
Mark M. Wilde (LSU) 76 / 113
![Page 305: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/305.jpg)
Evaluating quantum entropy
How do we evaluate the formula for quantum entropy of a state ρA?
Consider spectral decomposition:
ρA =∑x
pX (x)|x〉〈x |A.
Then, with η(x) = −x log2(x),
H(A)ρ = Tr{η(ρA)} = Tr
{∑x
η(pX (x))|x〉〈x |A
}=∑x
η(pX (x)) Tr{|x〉〈x |A} =∑x
η(pX (x)) = H(pX ).
Quantum entropy of ρA is equal to Shannon entropy of eigenvalues.
⇒ Entropy of a pure state is equal to zero.
Mark M. Wilde (LSU) 76 / 113
![Page 306: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/306.jpg)
Evaluating quantum entropy
How do we evaluate the formula for quantum entropy of a state ρA?
Consider spectral decomposition:
ρA =∑x
pX (x)|x〉〈x |A.
Then, with η(x) = −x log2(x),
H(A)ρ = Tr{η(ρA)} = Tr
{∑x
η(pX (x))|x〉〈x |A
}=∑x
η(pX (x)) Tr{|x〉〈x |A} =∑x
η(pX (x)) = H(pX ).
Quantum entropy of ρA is equal to Shannon entropy of eigenvalues.
⇒ Entropy of a pure state is equal to zero.
Mark M. Wilde (LSU) 76 / 113
![Page 307: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/307.jpg)
Bipartite pure-state entanglement
Let |ψ〉〈ψ|AB be a pure state.
By Schmidt decomposition theorem, we know that
|ψ〉AB =∑x
√pX (x)|x〉A ⊗ |x〉B ,
for prob. distribution pX and orthonormal bases {|x〉A} and {|x〉B}.
⇒ Eigenvalues of marginal states TrB{|ψ〉〈ψ|AB} and TrA{|ψ〉〈ψ|AB}are equal.
Thus, H(A)ρ = H(B)ρ if ρAB is a pure state.
Exercise: For a tripartite pure state |φ〉〈φ|ABC ,
H(A|B)φ + H(A|C )φ = 0.
Mark M. Wilde (LSU) 77 / 113
![Page 308: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/308.jpg)
Bipartite pure-state entanglement
Let |ψ〉〈ψ|AB be a pure state.
By Schmidt decomposition theorem, we know that
|ψ〉AB =∑x
√pX (x)|x〉A ⊗ |x〉B ,
for prob. distribution pX and orthonormal bases {|x〉A} and {|x〉B}.
⇒ Eigenvalues of marginal states TrB{|ψ〉〈ψ|AB} and TrA{|ψ〉〈ψ|AB}are equal.
Thus, H(A)ρ = H(B)ρ if ρAB is a pure state.
Exercise: For a tripartite pure state |φ〉〈φ|ABC ,
H(A|B)φ + H(A|C )φ = 0.
Mark M. Wilde (LSU) 77 / 113
![Page 309: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/309.jpg)
Bipartite pure-state entanglement
Let |ψ〉〈ψ|AB be a pure state.
By Schmidt decomposition theorem, we know that
|ψ〉AB =∑x
√pX (x)|x〉A ⊗ |x〉B ,
for prob. distribution pX and orthonormal bases {|x〉A} and {|x〉B}.
⇒ Eigenvalues of marginal states TrB{|ψ〉〈ψ|AB} and TrA{|ψ〉〈ψ|AB}are equal.
Thus, H(A)ρ = H(B)ρ if ρAB is a pure state.
Exercise: For a tripartite pure state |φ〉〈φ|ABC ,
H(A|B)φ + H(A|C )φ = 0.
Mark M. Wilde (LSU) 77 / 113
![Page 310: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/310.jpg)
Bipartite pure-state entanglement
Let |ψ〉〈ψ|AB be a pure state.
By Schmidt decomposition theorem, we know that
|ψ〉AB =∑x
√pX (x)|x〉A ⊗ |x〉B ,
for prob. distribution pX and orthonormal bases {|x〉A} and {|x〉B}.
⇒ Eigenvalues of marginal states TrB{|ψ〉〈ψ|AB} and TrA{|ψ〉〈ψ|AB}are equal.
Thus, H(A)ρ = H(B)ρ if ρAB is a pure state.
Exercise: For a tripartite pure state |φ〉〈φ|ABC ,
H(A|B)φ + H(A|C )φ = 0.
Mark M. Wilde (LSU) 77 / 113
![Page 311: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/311.jpg)
Bipartite pure-state entanglement
Let |ψ〉〈ψ|AB be a pure state.
By Schmidt decomposition theorem, we know that
|ψ〉AB =∑x
√pX (x)|x〉A ⊗ |x〉B ,
for prob. distribution pX and orthonormal bases {|x〉A} and {|x〉B}.
⇒ Eigenvalues of marginal states TrB{|ψ〉〈ψ|AB} and TrA{|ψ〉〈ψ|AB}are equal.
Thus, H(A)ρ = H(B)ρ if ρAB is a pure state.
Exercise: For a tripartite pure state |φ〉〈φ|ABC ,
H(A|B)φ + H(A|C )φ = 0.
Mark M. Wilde (LSU) 77 / 113
![Page 312: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/312.jpg)
Bipartite pure-state entanglement
Let |ψ〉〈ψ|AB be a pure state.
By Schmidt decomposition theorem, we know that
|ψ〉AB =∑x
√pX (x)|x〉A ⊗ |x〉B ,
for prob. distribution pX and orthonormal bases {|x〉A} and {|x〉B}.
⇒ Eigenvalues of marginal states TrB{|ψ〉〈ψ|AB} and TrA{|ψ〉〈ψ|AB}are equal.
Thus, H(A)ρ = H(B)ρ if ρAB is a pure state.
Exercise: For a tripartite pure state |φ〉〈φ|ABC ,
H(A|B)φ + H(A|C )φ = 0.
Mark M. Wilde (LSU) 77 / 113
![Page 313: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/313.jpg)
Conditional quantum entropy can be negative
One of the most striking differences between classical and quantuminformation theory: conditional quantum entropy can be negative.
Consider the conditional quantum entropy of the ebit |Φ〉〈Φ|AB .
The global state is pure, while the marginal TrA{|Φ〉〈Φ|AB} ismaximally mixed.
This implies that H(AB)Φ = 0 and H(B)Φ = 1, and thus
H(A|B)Φ = −1.
If a state σAB is separable, then one can show that H(A|B)σ ≥ 0. Soa negative conditional entropy implies that a state is entangled(signature of entanglement).
Mark M. Wilde (LSU) 78 / 113
![Page 314: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/314.jpg)
Conditional quantum entropy can be negative
One of the most striking differences between classical and quantuminformation theory: conditional quantum entropy can be negative.
Consider the conditional quantum entropy of the ebit |Φ〉〈Φ|AB .
The global state is pure, while the marginal TrA{|Φ〉〈Φ|AB} ismaximally mixed.
This implies that H(AB)Φ = 0 and H(B)Φ = 1, and thus
H(A|B)Φ = −1.
If a state σAB is separable, then one can show that H(A|B)σ ≥ 0. Soa negative conditional entropy implies that a state is entangled(signature of entanglement).
Mark M. Wilde (LSU) 78 / 113
![Page 315: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/315.jpg)
Conditional quantum entropy can be negative
One of the most striking differences between classical and quantuminformation theory: conditional quantum entropy can be negative.
Consider the conditional quantum entropy of the ebit |Φ〉〈Φ|AB .
The global state is pure, while the marginal TrA{|Φ〉〈Φ|AB} ismaximally mixed.
This implies that H(AB)Φ = 0 and H(B)Φ = 1, and thus
H(A|B)Φ = −1.
If a state σAB is separable, then one can show that H(A|B)σ ≥ 0. Soa negative conditional entropy implies that a state is entangled(signature of entanglement).
Mark M. Wilde (LSU) 78 / 113
![Page 316: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/316.jpg)
Conditional quantum entropy can be negative
One of the most striking differences between classical and quantuminformation theory: conditional quantum entropy can be negative.
Consider the conditional quantum entropy of the ebit |Φ〉〈Φ|AB .
The global state is pure, while the marginal TrA{|Φ〉〈Φ|AB} ismaximally mixed.
This implies that H(AB)Φ = 0 and H(B)Φ = 1, and thus
H(A|B)Φ = −1.
If a state σAB is separable, then one can show that H(A|B)σ ≥ 0. Soa negative conditional entropy implies that a state is entangled(signature of entanglement).
Mark M. Wilde (LSU) 78 / 113
![Page 317: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/317.jpg)
Conditional quantum entropy can be negative
One of the most striking differences between classical and quantuminformation theory: conditional quantum entropy can be negative.
Consider the conditional quantum entropy of the ebit |Φ〉〈Φ|AB .
The global state is pure, while the marginal TrA{|Φ〉〈Φ|AB} ismaximally mixed.
This implies that H(AB)Φ = 0 and H(B)Φ = 1, and thus
H(A|B)Φ = −1.
If a state σAB is separable, then one can show that H(A|B)σ ≥ 0. Soa negative conditional entropy implies that a state is entangled(signature of entanglement).
Mark M. Wilde (LSU) 78 / 113
![Page 318: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/318.jpg)
Conditional quantum entropy can be negative
One of the most striking differences between classical and quantuminformation theory: conditional quantum entropy can be negative.
Consider the conditional quantum entropy of the ebit |Φ〉〈Φ|AB .
The global state is pure, while the marginal TrA{|Φ〉〈Φ|AB} ismaximally mixed.
This implies that H(AB)Φ = 0 and H(B)Φ = 1, and thus
H(A|B)Φ = −1.
If a state σAB is separable, then one can show that H(A|B)σ ≥ 0. Soa negative conditional entropy implies that a state is entangled(signature of entanglement).
Mark M. Wilde (LSU) 78 / 113
![Page 319: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/319.jpg)
Strong subadditivity
Strong subadditivity
Let ρABC be a tripartite quantum state. Then
I (A;B|C )ρ ≥ 0.
Equivalent statements (by definition)
Entropy sum of two individual systems is larger than entropy sum oftheir union and intersection:
H(AC )ρ + H(BC )ρ ≥ H(ABC )ρ + H(C )ρ.
Conditional entropy does not decrease under the loss of system A:
H(B|C )ρ ≥ H(B|AC )ρ.
Mark M. Wilde (LSU) 79 / 113
![Page 320: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/320.jpg)
Strong subadditivity
Strong subadditivity
Let ρABC be a tripartite quantum state. Then
I (A;B|C )ρ ≥ 0.
Equivalent statements (by definition)
Entropy sum of two individual systems is larger than entropy sum oftheir union and intersection:
H(AC )ρ + H(BC )ρ ≥ H(ABC )ρ + H(C )ρ.
Conditional entropy does not decrease under the loss of system A:
H(B|C )ρ ≥ H(B|AC )ρ.
Mark M. Wilde (LSU) 79 / 113
![Page 321: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/321.jpg)
Strong subadditivity
Strong subadditivity
Let ρABC be a tripartite quantum state. Then
I (A;B|C )ρ ≥ 0.
Equivalent statements (by definition)
Entropy sum of two individual systems is larger than entropy sum oftheir union and intersection:
H(AC )ρ + H(BC )ρ ≥ H(ABC )ρ + H(C )ρ.
Conditional entropy does not decrease under the loss of system A:
H(B|C )ρ ≥ H(B|AC )ρ.
Mark M. Wilde (LSU) 79 / 113
![Page 322: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/322.jpg)
Strong subadditivity
Strong subadditivity
Let ρABC be a tripartite quantum state. Then
I (A;B|C )ρ ≥ 0.
Equivalent statements (by definition)
Entropy sum of two individual systems is larger than entropy sum oftheir union and intersection:
H(AC )ρ + H(BC )ρ ≥ H(ABC )ρ + H(C )ρ.
Conditional entropy does not decrease under the loss of system A:
H(B|C )ρ ≥ H(B|AC )ρ.
Mark M. Wilde (LSU) 79 / 113
![Page 323: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/323.jpg)
Strong subadditivity
Strong subadditivity
Let ρABC be a tripartite quantum state. Then
I (A;B|C )ρ ≥ 0.
Equivalent statements (by definition)
Entropy sum of two individual systems is larger than entropy sum oftheir union and intersection:
H(AC )ρ + H(BC )ρ ≥ H(ABC )ρ + H(C )ρ.
Conditional entropy does not decrease under the loss of system A:
H(B|C )ρ ≥ H(B|AC )ρ.
Mark M. Wilde (LSU) 79 / 113
![Page 324: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/324.jpg)
Monogamy of entanglement
By employing strong subadditivity and the Schmidt decomposition,we see that
H(A|B)ρ + H(A|C )ρ ≥ 0.
This is a nontrivial statement for quantum states, given that H(A|B)ρcan be negative.
Thus, if H(A|B)ρ < 0, implying that Alice is entangled with Bob,then it must be the case that H(A|C )ρ is large enough such that thesum is non-negative.
Often called “monogamy of entanglement,” because it says that Alicecannot be strongly entangled with both Bob and Charlie.
Mark M. Wilde (LSU) 80 / 113
![Page 325: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/325.jpg)
Monogamy of entanglement
By employing strong subadditivity and the Schmidt decomposition,we see that
H(A|B)ρ + H(A|C )ρ ≥ 0.
This is a nontrivial statement for quantum states, given that H(A|B)ρcan be negative.
Thus, if H(A|B)ρ < 0, implying that Alice is entangled with Bob,then it must be the case that H(A|C )ρ is large enough such that thesum is non-negative.
Often called “monogamy of entanglement,” because it says that Alicecannot be strongly entangled with both Bob and Charlie.
Mark M. Wilde (LSU) 80 / 113
![Page 326: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/326.jpg)
Monogamy of entanglement
By employing strong subadditivity and the Schmidt decomposition,we see that
H(A|B)ρ + H(A|C )ρ ≥ 0.
This is a nontrivial statement for quantum states, given that H(A|B)ρcan be negative.
Thus, if H(A|B)ρ < 0, implying that Alice is entangled with Bob,then it must be the case that H(A|C )ρ is large enough such that thesum is non-negative.
Often called “monogamy of entanglement,” because it says that Alicecannot be strongly entangled with both Bob and Charlie.
Mark M. Wilde (LSU) 80 / 113
![Page 327: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/327.jpg)
Monogamy of entanglement
By employing strong subadditivity and the Schmidt decomposition,we see that
H(A|B)ρ + H(A|C )ρ ≥ 0.
This is a nontrivial statement for quantum states, given that H(A|B)ρcan be negative.
Thus, if H(A|B)ρ < 0, implying that Alice is entangled with Bob,then it must be the case that H(A|C )ρ is large enough such that thesum is non-negative.
Often called “monogamy of entanglement,” because it says that Alicecannot be strongly entangled with both Bob and Charlie.
Mark M. Wilde (LSU) 80 / 113
![Page 328: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/328.jpg)
Monogamy of entanglement
By employing strong subadditivity and the Schmidt decomposition,we see that
H(A|B)ρ + H(A|C )ρ ≥ 0.
This is a nontrivial statement for quantum states, given that H(A|B)ρcan be negative.
Thus, if H(A|B)ρ < 0, implying that Alice is entangled with Bob,then it must be the case that H(A|C )ρ is large enough such that thesum is non-negative.
Often called “monogamy of entanglement,” because it says that Alicecannot be strongly entangled with both Bob and Charlie.
Mark M. Wilde (LSU) 80 / 113
![Page 329: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/329.jpg)
Quantum data compression
Mark M. Wilde (LSU) 81 / 113
![Page 330: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/330.jpg)
Quantum information source
We model a quantum information source as an ensemble of purestates: {pX (x), |φx〉〈φx |}.
The source has expected density matrix
ρ =∑x
pX (x)|φx〉〈φx |. (3)
Every density matrix has a spectral decomposition:
ρ =∑z
pZ (z)|z〉〈z |,
where pZ is a probability distribution and {|z〉} is an O.N. basis. Thisdecomposition in general is different from the one in (3).
Mark M. Wilde (LSU) 82 / 113
![Page 331: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/331.jpg)
Quantum information source
We model a quantum information source as an ensemble of purestates: {pX (x), |φx〉〈φx |}.
The source has expected density matrix
ρ =∑x
pX (x)|φx〉〈φx |. (3)
Every density matrix has a spectral decomposition:
ρ =∑z
pZ (z)|z〉〈z |,
where pZ is a probability distribution and {|z〉} is an O.N. basis. Thisdecomposition in general is different from the one in (3).
Mark M. Wilde (LSU) 82 / 113
![Page 332: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/332.jpg)
Quantum information source
We model a quantum information source as an ensemble of purestates: {pX (x), |φx〉〈φx |}.
The source has expected density matrix
ρ =∑x
pX (x)|φx〉〈φx |. (3)
Every density matrix has a spectral decomposition:
ρ =∑z
pZ (z)|z〉〈z |,
where pZ is a probability distribution and {|z〉} is an O.N. basis. Thisdecomposition in general is different from the one in (3).
Mark M. Wilde (LSU) 82 / 113
![Page 333: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/333.jpg)
Quantum information source
We model a quantum information source as an ensemble of purestates: {pX (x), |φx〉〈φx |}.
The source has expected density matrix
ρ =∑x
pX (x)|φx〉〈φx |. (3)
Every density matrix has a spectral decomposition:
ρ =∑z
pZ (z)|z〉〈z |,
where pZ is a probability distribution and {|z〉} is an O.N. basis. Thisdecomposition in general is different from the one in (3).
Mark M. Wilde (LSU) 82 / 113
![Page 334: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/334.jpg)
Quantum data compression protocols
Inspired by Shannon, we consider independent calls of the quantuminformation source and allow for compression schemes that haveslight error which vanishes in the limit of many calls of the source.
An (n,R, ε) quantum data compression scheme consists of anencoding channel En, with output system W , and a decoding channelDn such that
1
nlog2 dim(HW ) ≤ R,
and ∑xn
pX n(xn)F (|φxn〉〈φxn |, (Dn ◦ En)[|φxn〉〈φxn |]) ≥ 1− ε.
A rate R is achievable if for all ε ∈ (0, 1) and sufficiently large n,there exists an (n,R, ε) quantum compression scheme.
Quantum data compression limit = infimum of achievable rates.
Mark M. Wilde (LSU) 83 / 113
![Page 335: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/335.jpg)
Quantum data compression protocols
Inspired by Shannon, we consider independent calls of the quantuminformation source and allow for compression schemes that haveslight error which vanishes in the limit of many calls of the source.
An (n,R, ε) quantum data compression scheme consists of anencoding channel En, with output system W , and a decoding channelDn such that
1
nlog2 dim(HW ) ≤ R,
and ∑xn
pX n(xn)F (|φxn〉〈φxn |, (Dn ◦ En)[|φxn〉〈φxn |]) ≥ 1− ε.
A rate R is achievable if for all ε ∈ (0, 1) and sufficiently large n,there exists an (n,R, ε) quantum compression scheme.
Quantum data compression limit = infimum of achievable rates.
Mark M. Wilde (LSU) 83 / 113
![Page 336: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/336.jpg)
Quantum data compression protocols
Inspired by Shannon, we consider independent calls of the quantuminformation source and allow for compression schemes that haveslight error which vanishes in the limit of many calls of the source.
An (n,R, ε) quantum data compression scheme consists of anencoding channel En, with output system W , and a decoding channelDn such that
1
nlog2 dim(HW ) ≤ R,
and ∑xn
pX n(xn)F (|φxn〉〈φxn |, (Dn ◦ En)[|φxn〉〈φxn |]) ≥ 1− ε.
A rate R is achievable if for all ε ∈ (0, 1) and sufficiently large n,there exists an (n,R, ε) quantum compression scheme.
Quantum data compression limit = infimum of achievable rates.
Mark M. Wilde (LSU) 83 / 113
![Page 337: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/337.jpg)
Quantum data compression protocols
Inspired by Shannon, we consider independent calls of the quantuminformation source and allow for compression schemes that haveslight error which vanishes in the limit of many calls of the source.
An (n,R, ε) quantum data compression scheme consists of anencoding channel En, with output system W , and a decoding channelDn such that
1
nlog2 dim(HW ) ≤ R,
and ∑xn
pX n(xn)F (|φxn〉〈φxn |, (Dn ◦ En)[|φxn〉〈φxn |]) ≥ 1− ε.
A rate R is achievable if for all ε ∈ (0, 1) and sufficiently large n,there exists an (n,R, ε) quantum compression scheme.
Quantum data compression limit = infimum of achievable rates.
Mark M. Wilde (LSU) 83 / 113
![Page 338: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/338.jpg)
Quantum data compression protocols
Inspired by Shannon, we consider independent calls of the quantuminformation source and allow for compression schemes that haveslight error which vanishes in the limit of many calls of the source.
An (n,R, ε) quantum data compression scheme consists of anencoding channel En, with output system W , and a decoding channelDn such that
1
nlog2 dim(HW ) ≤ R,
and ∑xn
pX n(xn)F (|φxn〉〈φxn |, (Dn ◦ En)[|φxn〉〈φxn |]) ≥ 1− ε.
A rate R is achievable if for all ε ∈ (0, 1) and sufficiently large n,there exists an (n,R, ε) quantum compression scheme.
Quantum data compression limit = infimum of achievable rates.
Mark M. Wilde (LSU) 83 / 113
![Page 339: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/339.jpg)
Quantum data compression protocols
Inspired by Shannon, we consider independent calls of the quantuminformation source and allow for compression schemes that haveslight error which vanishes in the limit of many calls of the source.
An (n,R, ε) quantum data compression scheme consists of anencoding channel En, with output system W , and a decoding channelDn such that
1
nlog2 dim(HW ) ≤ R,
and ∑xn
pX n(xn)F (|φxn〉〈φxn |, (Dn ◦ En)[|φxn〉〈φxn |]) ≥ 1− ε.
A rate R is achievable if for all ε ∈ (0, 1) and sufficiently large n,there exists an (n,R, ε) quantum compression scheme.
Quantum data compression limit = infimum of achievable rates.
Mark M. Wilde (LSU) 83 / 113
![Page 340: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/340.jpg)
Quantum data compression theorem
The quantum data compression limit of a source {pX (x), |φx〉〈φx |} isequal to the quantum entropy of ρ =
∑x pX (x)|φx〉〈φx |.
Focus on achievability part. To prove it, we use the notion ofquantum typicality.
Mark M. Wilde (LSU) 84 / 113
![Page 341: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/341.jpg)
Quantum data compression theorem
The quantum data compression limit of a source {pX (x), |φx〉〈φx |} isequal to the quantum entropy of ρ =
∑x pX (x)|φx〉〈φx |.
Focus on achievability part. To prove it, we use the notion ofquantum typicality.
Mark M. Wilde (LSU) 84 / 113
![Page 342: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/342.jpg)
Quantum data compression theorem
The quantum data compression limit of a source {pX (x), |φx〉〈φx |} isequal to the quantum entropy of ρ =
∑x pX (x)|φx〉〈φx |.
Focus on achievability part. To prove it, we use the notion ofquantum typicality.
Mark M. Wilde (LSU) 84 / 113
![Page 343: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/343.jpg)
Quantum typicality
Given a density matrix ρ with spectral decomposition∑
z pZ (z)|z〉〈z |,define its (n, δ)-typical subspace by
T ρn,δ ≡ span
{|zn〉 :
∣∣∣∣−1
nlog2 pZn(zn)− H(ρ)
∣∣∣∣ ≤ δ} , where
pZn(zn) ≡ pZ (z1) · · · pZ (zn), |zn〉 ≡ |z1〉 ⊗ · · · ⊗ |zn〉.
Let Πρn,δ denote the projection onto T ρ
n,δ.
Then,
Tr{Πρn,δρ
⊗n} ≥ 1− ε,
(1− ε)2n[H(ρ)−δ] ≤ Tr{Πρn,δ} ≤ 2n[H(ρ)+δ],
2−n[H(ρ)+δ]Πρn,δ ≤ Πρ
n,δρ⊗nΠρ
n,δ ≤ 2−n[H(ρ)−δ]Πρn,δ.
Inequalities with ε are true for all ε ∈ (0, 1) and sufficiently large n.
Mark M. Wilde (LSU) 85 / 113
![Page 344: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/344.jpg)
Quantum typicality
Given a density matrix ρ with spectral decomposition∑
z pZ (z)|z〉〈z |,define its (n, δ)-typical subspace by
T ρn,δ ≡ span
{|zn〉 :
∣∣∣∣−1
nlog2 pZn(zn)− H(ρ)
∣∣∣∣ ≤ δ} , where
pZn(zn) ≡ pZ (z1) · · · pZ (zn), |zn〉 ≡ |z1〉 ⊗ · · · ⊗ |zn〉.
Let Πρn,δ denote the projection onto T ρ
n,δ.
Then,
Tr{Πρn,δρ
⊗n} ≥ 1− ε,
(1− ε)2n[H(ρ)−δ] ≤ Tr{Πρn,δ} ≤ 2n[H(ρ)+δ],
2−n[H(ρ)+δ]Πρn,δ ≤ Πρ
n,δρ⊗nΠρ
n,δ ≤ 2−n[H(ρ)−δ]Πρn,δ.
Inequalities with ε are true for all ε ∈ (0, 1) and sufficiently large n.
Mark M. Wilde (LSU) 85 / 113
![Page 345: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/345.jpg)
Quantum typicality
Given a density matrix ρ with spectral decomposition∑
z pZ (z)|z〉〈z |,define its (n, δ)-typical subspace by
T ρn,δ ≡ span
{|zn〉 :
∣∣∣∣−1
nlog2 pZn(zn)− H(ρ)
∣∣∣∣ ≤ δ} , where
pZn(zn) ≡ pZ (z1) · · · pZ (zn), |zn〉 ≡ |z1〉 ⊗ · · · ⊗ |zn〉.
Let Πρn,δ denote the projection onto T ρ
n,δ.
Then,
Tr{Πρn,δρ
⊗n} ≥ 1− ε,
(1− ε)2n[H(ρ)−δ] ≤ Tr{Πρn,δ} ≤ 2n[H(ρ)+δ],
2−n[H(ρ)+δ]Πρn,δ ≤ Πρ
n,δρ⊗nΠρ
n,δ ≤ 2−n[H(ρ)−δ]Πρn,δ.
Inequalities with ε are true for all ε ∈ (0, 1) and sufficiently large n.
Mark M. Wilde (LSU) 85 / 113
![Page 346: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/346.jpg)
Quantum typicality
Given a density matrix ρ with spectral decomposition∑
z pZ (z)|z〉〈z |,define its (n, δ)-typical subspace by
T ρn,δ ≡ span
{|zn〉 :
∣∣∣∣−1
nlog2 pZn(zn)− H(ρ)
∣∣∣∣ ≤ δ} , where
pZn(zn) ≡ pZ (z1) · · · pZ (zn), |zn〉 ≡ |z1〉 ⊗ · · · ⊗ |zn〉.
Let Πρn,δ denote the projection onto T ρ
n,δ.
Then,
Tr{Πρn,δρ
⊗n} ≥ 1− ε,
(1− ε)2n[H(ρ)−δ] ≤ Tr{Πρn,δ} ≤ 2n[H(ρ)+δ],
2−n[H(ρ)+δ]Πρn,δ ≤ Πρ
n,δρ⊗nΠρ
n,δ ≤ 2−n[H(ρ)−δ]Πρn,δ.
Inequalities with ε are true for all ε ∈ (0, 1) and sufficiently large n.
Mark M. Wilde (LSU) 85 / 113
![Page 347: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/347.jpg)
Quantum typicality
Given a density matrix ρ with spectral decomposition∑
z pZ (z)|z〉〈z |,define its (n, δ)-typical subspace by
T ρn,δ ≡ span
{|zn〉 :
∣∣∣∣−1
nlog2 pZn(zn)− H(ρ)
∣∣∣∣ ≤ δ} , where
pZn(zn) ≡ pZ (z1) · · · pZ (zn), |zn〉 ≡ |z1〉 ⊗ · · · ⊗ |zn〉.
Let Πρn,δ denote the projection onto T ρ
n,δ.
Then,
Tr{Πρn,δρ
⊗n} ≥ 1− ε,
(1− ε)2n[H(ρ)−δ] ≤ Tr{Πρn,δ} ≤ 2n[H(ρ)+δ],
2−n[H(ρ)+δ]Πρn,δ ≤ Πρ
n,δρ⊗nΠρ
n,δ ≤ 2−n[H(ρ)−δ]Πρn,δ.
Inequalities with ε are true for all ε ∈ (0, 1) and sufficiently large n.
Mark M. Wilde (LSU) 85 / 113
![Page 348: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/348.jpg)
Quantum data compression
Main idea for quantum data compression: measure typical subspace.Successful with probability 1− ε.
If successful, perform a unitary that rotates typical subspace to spaceof dimension ≤ 2n[H(ρ)+δ] (represented with n[H(ρ) + δ] qubits).
Send qubits to Bob, who then undoes the compression unitary.
Scheme is guaranteed to meet the fidelity criterion.
Mark M. Wilde (LSU) 86 / 113
![Page 349: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/349.jpg)
Quantum data compression
Main idea for quantum data compression: measure typical subspace.Successful with probability 1− ε.
If successful, perform a unitary that rotates typical subspace to spaceof dimension ≤ 2n[H(ρ)+δ] (represented with n[H(ρ) + δ] qubits).
Send qubits to Bob, who then undoes the compression unitary.
Scheme is guaranteed to meet the fidelity criterion.
Mark M. Wilde (LSU) 86 / 113
![Page 350: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/350.jpg)
Quantum data compression
Main idea for quantum data compression: measure typical subspace.Successful with probability 1− ε.
If successful, perform a unitary that rotates typical subspace to spaceof dimension ≤ 2n[H(ρ)+δ] (represented with n[H(ρ) + δ] qubits).
Send qubits to Bob, who then undoes the compression unitary.
Scheme is guaranteed to meet the fidelity criterion.
Mark M. Wilde (LSU) 86 / 113
![Page 351: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/351.jpg)
Quantum data compression
Main idea for quantum data compression: measure typical subspace.Successful with probability 1− ε.
If successful, perform a unitary that rotates typical subspace to spaceof dimension ≤ 2n[H(ρ)+δ] (represented with n[H(ρ) + δ] qubits).
Send qubits to Bob, who then undoes the compression unitary.
Scheme is guaranteed to meet the fidelity criterion.
Mark M. Wilde (LSU) 86 / 113
![Page 352: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/352.jpg)
Quantum data compression
Main idea for quantum data compression: measure typical subspace.Successful with probability 1− ε.
If successful, perform a unitary that rotates typical subspace to spaceof dimension ≤ 2n[H(ρ)+δ] (represented with n[H(ρ) + δ] qubits).
Send qubits to Bob, who then undoes the compression unitary.
Scheme is guaranteed to meet the fidelity criterion.
Mark M. Wilde (LSU) 86 / 113
![Page 353: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/353.jpg)
Classical communication
Mark M. Wilde (LSU) 87 / 113
![Page 354: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/354.jpg)
Classical communication code
Suppose that Alice and Bob are connected by a quantum channelNA→B and that they are allowed to use it n times. The resultingchannel is N⊗nA→B , with Kraus operators that are tensor products ofthe individual Kraus operators.
An (n,R, ε) classical comm. code consists of an encoding channelEM′→An and a decoding measurement channel DBn→M̂ such that:
F (ΦMM̂ , (DBn→M̂ ◦ N⊗nA→B ◦ EM′→An)(ΦMM′)) ≥ 1− ε,
where
ΦMM̂ ≡1
dim(HM)
∑m
|m〉〈m|M ⊗ |m〉〈m|M̂ ,
and 1n log2(dim(HM)) ≥ R.
Note that ΦMM̂ represents a classical state, and the goal is for thecoding scheme to preserve the classical correlations in this state.
Mark M. Wilde (LSU) 88 / 113
![Page 355: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/355.jpg)
Classical communication code
Suppose that Alice and Bob are connected by a quantum channelNA→B and that they are allowed to use it n times. The resultingchannel is N⊗nA→B , with Kraus operators that are tensor products ofthe individual Kraus operators.
An (n,R, ε) classical comm. code consists of an encoding channelEM′→An and a decoding measurement channel DBn→M̂ such that:
F (ΦMM̂ , (DBn→M̂ ◦ N⊗nA→B ◦ EM′→An)(ΦMM′)) ≥ 1− ε,
where
ΦMM̂ ≡1
dim(HM)
∑m
|m〉〈m|M ⊗ |m〉〈m|M̂ ,
and 1n log2(dim(HM)) ≥ R.
Note that ΦMM̂ represents a classical state, and the goal is for thecoding scheme to preserve the classical correlations in this state.
Mark M. Wilde (LSU) 88 / 113
![Page 356: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/356.jpg)
Classical communication code
Suppose that Alice and Bob are connected by a quantum channelNA→B and that they are allowed to use it n times. The resultingchannel is N⊗nA→B , with Kraus operators that are tensor products ofthe individual Kraus operators.
An (n,R, ε) classical comm. code consists of an encoding channelEM′→An and a decoding measurement channel DBn→M̂ such that:
F (ΦMM̂ , (DBn→M̂ ◦ N⊗nA→B ◦ EM′→An)(ΦMM′)) ≥ 1− ε,
where
ΦMM̂ ≡1
dim(HM)
∑m
|m〉〈m|M ⊗ |m〉〈m|M̂ ,
and 1n log2(dim(HM)) ≥ R.
Note that ΦMM̂ represents a classical state, and the goal is for thecoding scheme to preserve the classical correlations in this state.
Mark M. Wilde (LSU) 88 / 113
![Page 357: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/357.jpg)
Classical communication code
Suppose that Alice and Bob are connected by a quantum channelNA→B and that they are allowed to use it n times. The resultingchannel is N⊗nA→B , with Kraus operators that are tensor products ofthe individual Kraus operators.
An (n,R, ε) classical comm. code consists of an encoding channelEM′→An and a decoding measurement channel DBn→M̂ such that:
F (ΦMM̂ , (DBn→M̂ ◦ N⊗nA→B ◦ EM′→An)(ΦMM′)) ≥ 1− ε,
where
ΦMM̂ ≡1
dim(HM)
∑m
|m〉〈m|M ⊗ |m〉〈m|M̂ ,
and 1n log2(dim(HM)) ≥ R.
Note that ΦMM̂ represents a classical state, and the goal is for thecoding scheme to preserve the classical correlations in this state.
Mark M. Wilde (LSU) 88 / 113
![Page 358: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/358.jpg)
Classical communication code
Suppose that Alice and Bob are connected by a quantum channelNA→B and that they are allowed to use it n times. The resultingchannel is N⊗nA→B , with Kraus operators that are tensor products ofthe individual Kraus operators.
An (n,R, ε) classical comm. code consists of an encoding channelEM′→An and a decoding measurement channel DBn→M̂ such that:
F (ΦMM̂ , (DBn→M̂ ◦ N⊗nA→B ◦ EM′→An)(ΦMM′)) ≥ 1− ε,
where
ΦMM̂ ≡1
dim(HM)
∑m
|m〉〈m|M ⊗ |m〉〈m|M̂ ,
and 1n log2(dim(HM)) ≥ R.
Note that ΦMM̂ represents a classical state, and the goal is for thecoding scheme to preserve the classical correlations in this state.
Mark M. Wilde (LSU) 88 / 113
![Page 359: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/359.jpg)
Schematic of a classical communication code
Mark M. Wilde (LSU) 89 / 113
![Page 360: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/360.jpg)
Classical capacity
A rate R for classical communication is achievable if for all ε ∈ (0, 1)and sufficiently large n, there exists an (n,R, ε) classicalcommunication code.
The classical capacity C (N ) of a quantum channel N is equal to thesupremum of all achievable rates.
Mark M. Wilde (LSU) 90 / 113
![Page 361: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/361.jpg)
Classical capacity
A rate R for classical communication is achievable if for all ε ∈ (0, 1)and sufficiently large n, there exists an (n,R, ε) classicalcommunication code.
The classical capacity C (N ) of a quantum channel N is equal to thesupremum of all achievable rates.
Mark M. Wilde (LSU) 90 / 113
![Page 362: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/362.jpg)
Classical capacity
A rate R for classical communication is achievable if for all ε ∈ (0, 1)and sufficiently large n, there exists an (n,R, ε) classicalcommunication code.
The classical capacity C (N ) of a quantum channel N is equal to thesupremum of all achievable rates.
Mark M. Wilde (LSU) 90 / 113
![Page 363: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/363.jpg)
What is known about classical capacity
Lower bound on classical capacity:
χ(N ) ≤ C (N )
where χ(N ) = maxpX (x),ρxA
I (X ;B)ω,
ωXB ≡∑x
pX (x)|x〉〈x |X ⊗N (ρxA).
For some special channels, we know that χ(N ) = C (N ).
But it is also known that there exists a channel for which
χ(N ) < C (N ).
This superadditivity phenomenon is due to quantum entanglement.
Mark M. Wilde (LSU) 91 / 113
![Page 364: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/364.jpg)
What is known about classical capacity
Lower bound on classical capacity:
χ(N ) ≤ C (N )
where χ(N ) = maxpX (x),ρxA
I (X ;B)ω,
ωXB ≡∑x
pX (x)|x〉〈x |X ⊗N (ρxA).
For some special channels, we know that χ(N ) = C (N ).
But it is also known that there exists a channel for which
χ(N ) < C (N ).
This superadditivity phenomenon is due to quantum entanglement.
Mark M. Wilde (LSU) 91 / 113
![Page 365: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/365.jpg)
What is known about classical capacity
Lower bound on classical capacity:
χ(N ) ≤ C (N )
where χ(N ) = maxpX (x),ρxA
I (X ;B)ω,
ωXB ≡∑x
pX (x)|x〉〈x |X ⊗N (ρxA).
For some special channels, we know that χ(N ) = C (N ).
But it is also known that there exists a channel for which
χ(N ) < C (N ).
This superadditivity phenomenon is due to quantum entanglement.
Mark M. Wilde (LSU) 91 / 113
![Page 366: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/366.jpg)
What is known about classical capacity
Lower bound on classical capacity:
χ(N ) ≤ C (N )
where χ(N ) = maxpX (x),ρxA
I (X ;B)ω,
ωXB ≡∑x
pX (x)|x〉〈x |X ⊗N (ρxA).
For some special channels, we know that χ(N ) = C (N ).
But it is also known that there exists a channel for which
χ(N ) < C (N ).
This superadditivity phenomenon is due to quantum entanglement.
Mark M. Wilde (LSU) 91 / 113
![Page 367: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/367.jpg)
Achievability part: Random coding
Borrow the idea of random coding from Shannon, but then we needto figure out a decoding channel.
Consider an ensemble {pX (x), ρxA} that Alice can pick at the channelinput. This leads to the output ensemble
{pX (x), σxA ≡ NA→B(ρxA)}.
So pick classical codewords randomly according to pX (x). This leadsto a codebook {xn(m) ≡ x1(m) · · · xn(m)}m∈[dim(HM)].
The channel output after sending the mth message is
σxn(m)Bn ≡ σx1(m)
B1⊗ · · · ⊗ σxn(m)
Bn.
Mark M. Wilde (LSU) 92 / 113
![Page 368: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/368.jpg)
Achievability part: Random coding
Borrow the idea of random coding from Shannon, but then we needto figure out a decoding channel.
Consider an ensemble {pX (x), ρxA} that Alice can pick at the channelinput. This leads to the output ensemble
{pX (x), σxA ≡ NA→B(ρxA)}.
So pick classical codewords randomly according to pX (x). This leadsto a codebook {xn(m) ≡ x1(m) · · · xn(m)}m∈[dim(HM)].
The channel output after sending the mth message is
σxn(m)Bn ≡ σx1(m)
B1⊗ · · · ⊗ σxn(m)
Bn.
Mark M. Wilde (LSU) 92 / 113
![Page 369: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/369.jpg)
Achievability part: Random coding
Borrow the idea of random coding from Shannon, but then we needto figure out a decoding channel.
Consider an ensemble {pX (x), ρxA} that Alice can pick at the channelinput. This leads to the output ensemble
{pX (x), σxA ≡ NA→B(ρxA)}.
So pick classical codewords randomly according to pX (x). This leadsto a codebook {xn(m) ≡ x1(m) · · · xn(m)}m∈[dim(HM)].
The channel output after sending the mth message is
σxn(m)Bn ≡ σx1(m)
B1⊗ · · · ⊗ σxn(m)
Bn.
Mark M. Wilde (LSU) 92 / 113
![Page 370: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/370.jpg)
Achievability part: Random coding
Borrow the idea of random coding from Shannon, but then we needto figure out a decoding channel.
Consider an ensemble {pX (x), ρxA} that Alice can pick at the channelinput. This leads to the output ensemble
{pX (x), σxA ≡ NA→B(ρxA)}.
So pick classical codewords randomly according to pX (x). This leadsto a codebook {xn(m) ≡ x1(m) · · · xn(m)}m∈[dim(HM)].
The channel output after sending the mth message is
σxn(m)Bn ≡ σx1(m)
B1⊗ · · · ⊗ σxn(m)
Bn.
Mark M. Wilde (LSU) 92 / 113
![Page 371: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/371.jpg)
Achievability part: Random coding
Borrow the idea of random coding from Shannon, but then we needto figure out a decoding channel.
Consider an ensemble {pX (x), ρxA} that Alice can pick at the channelinput. This leads to the output ensemble
{pX (x), σxA ≡ NA→B(ρxA)}.
So pick classical codewords randomly according to pX (x). This leadsto a codebook {xn(m) ≡ x1(m) · · · xn(m)}m∈[dim(HM)].
The channel output after sending the mth message is
σxn(m)Bn ≡ σx1(m)
B1⊗ · · · ⊗ σxn(m)
Bn.
Mark M. Wilde (LSU) 92 / 113
![Page 372: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/372.jpg)
Achievability part: Sequential decoding
To every channel output σxn(m)Bn , there exists a conditionally typical
projector Πm, with properties similar to those of the typical projector.
A sequential decoding strategy consists of performing a sequence ofbinary tests using conditionally typical projectors, asking “Is it thefirst message? Is it the second message? etc.” until there is a “hit.”
When sending the mth message, the success probability in decoding itusing this strategy is
Tr{ΠmΠ̂m−1 · · · Π̂1σxn(m)Bn Π̂1 · · · Π̂m−1Πm},
where Π̂i ≡ I − Πi .
This implies that the error probability is
1− Tr{ΠmΠ̂m−1 · · · Π̂1σxn(m)Bn Π̂1 · · · Π̂m−1Πm}.
Mark M. Wilde (LSU) 93 / 113
![Page 373: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/373.jpg)
Achievability part: Sequential decoding
To every channel output σxn(m)Bn , there exists a conditionally typical
projector Πm, with properties similar to those of the typical projector.
A sequential decoding strategy consists of performing a sequence ofbinary tests using conditionally typical projectors, asking “Is it thefirst message? Is it the second message? etc.” until there is a “hit.”
When sending the mth message, the success probability in decoding itusing this strategy is
Tr{ΠmΠ̂m−1 · · · Π̂1σxn(m)Bn Π̂1 · · · Π̂m−1Πm},
where Π̂i ≡ I − Πi .
This implies that the error probability is
1− Tr{ΠmΠ̂m−1 · · · Π̂1σxn(m)Bn Π̂1 · · · Π̂m−1Πm}.
Mark M. Wilde (LSU) 93 / 113
![Page 374: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/374.jpg)
Achievability part: Sequential decoding
To every channel output σxn(m)Bn , there exists a conditionally typical
projector Πm, with properties similar to those of the typical projector.
A sequential decoding strategy consists of performing a sequence ofbinary tests using conditionally typical projectors, asking “Is it thefirst message? Is it the second message? etc.” until there is a “hit.”
When sending the mth message, the success probability in decoding itusing this strategy is
Tr{ΠmΠ̂m−1 · · · Π̂1σxn(m)Bn Π̂1 · · · Π̂m−1Πm},
where Π̂i ≡ I − Πi .
This implies that the error probability is
1− Tr{ΠmΠ̂m−1 · · · Π̂1σxn(m)Bn Π̂1 · · · Π̂m−1Πm}.
Mark M. Wilde (LSU) 93 / 113
![Page 375: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/375.jpg)
Achievability part: Sequential decoding
To every channel output σxn(m)Bn , there exists a conditionally typical
projector Πm, with properties similar to those of the typical projector.
A sequential decoding strategy consists of performing a sequence ofbinary tests using conditionally typical projectors, asking “Is it thefirst message? Is it the second message? etc.” until there is a “hit.”
When sending the mth message, the success probability in decoding itusing this strategy is
Tr{ΠmΠ̂m−1 · · · Π̂1σxn(m)Bn Π̂1 · · · Π̂m−1Πm},
where Π̂i ≡ I − Πi .
This implies that the error probability is
1− Tr{ΠmΠ̂m−1 · · · Π̂1σxn(m)Bn Π̂1 · · · Π̂m−1Πm}.
Mark M. Wilde (LSU) 93 / 113
![Page 376: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/376.jpg)
Achievability part: Sequential decoding
To every channel output σxn(m)Bn , there exists a conditionally typical
projector Πm, with properties similar to those of the typical projector.
A sequential decoding strategy consists of performing a sequence ofbinary tests using conditionally typical projectors, asking “Is it thefirst message? Is it the second message? etc.” until there is a “hit.”
When sending the mth message, the success probability in decoding itusing this strategy is
Tr{ΠmΠ̂m−1 · · · Π̂1σxn(m)Bn Π̂1 · · · Π̂m−1Πm},
where Π̂i ≡ I − Πi .
This implies that the error probability is
1− Tr{ΠmΠ̂m−1 · · · Π̂1σxn(m)Bn Π̂1 · · · Π̂m−1Πm}.
Mark M. Wilde (LSU) 93 / 113
![Page 377: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/377.jpg)
Error Analysis
The expected channel output with respect to the code distribution isσB =
∑x pX (x)σxB , which has a typical projection Πσ.
The error probability will ultimately change just slightly byincorporating this projection into the analysis:
Tr{Πσσxn(m)Bn Πσ}−Tr{ΠmΠ̂m−1 · · · Π̂1Πσσ
xn(m)Bn ΠσΠ̂1 · · · Π̂m−1Πm}.
Using a quantum version of the union bound, this can be boundedfrom above by
2
√√√√Tr{(I − Πm)Πσσxn(m)Bn Πσ}+
m−1∑i=1
Tr{ΠiΠσσxn(m)Bn Πσ}
The two terms above are exactly analogous to similar error terms thatarise in the analysis of Shannon’s channel coding theorem.
By taking an expecation with respect to the code distribution, we canthen analyze this error.
Mark M. Wilde (LSU) 94 / 113
![Page 378: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/378.jpg)
Error Analysis
The expected channel output with respect to the code distribution isσB =
∑x pX (x)σxB , which has a typical projection Πσ.
The error probability will ultimately change just slightly byincorporating this projection into the analysis:
Tr{Πσσxn(m)Bn Πσ}−Tr{ΠmΠ̂m−1 · · · Π̂1Πσσ
xn(m)Bn ΠσΠ̂1 · · · Π̂m−1Πm}.
Using a quantum version of the union bound, this can be boundedfrom above by
2
√√√√Tr{(I − Πm)Πσσxn(m)Bn Πσ}+
m−1∑i=1
Tr{ΠiΠσσxn(m)Bn Πσ}
The two terms above are exactly analogous to similar error terms thatarise in the analysis of Shannon’s channel coding theorem.
By taking an expecation with respect to the code distribution, we canthen analyze this error.
Mark M. Wilde (LSU) 94 / 113
![Page 379: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/379.jpg)
Error Analysis
The expected channel output with respect to the code distribution isσB =
∑x pX (x)σxB , which has a typical projection Πσ.
The error probability will ultimately change just slightly byincorporating this projection into the analysis:
Tr{Πσσxn(m)Bn Πσ}−Tr{ΠmΠ̂m−1 · · · Π̂1Πσσ
xn(m)Bn ΠσΠ̂1 · · · Π̂m−1Πm}.
Using a quantum version of the union bound, this can be boundedfrom above by
2
√√√√Tr{(I − Πm)Πσσxn(m)Bn Πσ}+
m−1∑i=1
Tr{ΠiΠσσxn(m)Bn Πσ}
The two terms above are exactly analogous to similar error terms thatarise in the analysis of Shannon’s channel coding theorem.
By taking an expecation with respect to the code distribution, we canthen analyze this error.
Mark M. Wilde (LSU) 94 / 113
![Page 380: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/380.jpg)
Error Analysis
The expected channel output with respect to the code distribution isσB =
∑x pX (x)σxB , which has a typical projection Πσ.
The error probability will ultimately change just slightly byincorporating this projection into the analysis:
Tr{Πσσxn(m)Bn Πσ}−Tr{ΠmΠ̂m−1 · · · Π̂1Πσσ
xn(m)Bn ΠσΠ̂1 · · · Π̂m−1Πm}.
Using a quantum version of the union bound, this can be boundedfrom above by
2
√√√√Tr{(I − Πm)Πσσxn(m)Bn Πσ}+
m−1∑i=1
Tr{ΠiΠσσxn(m)Bn Πσ}
The two terms above are exactly analogous to similar error terms thatarise in the analysis of Shannon’s channel coding theorem.
By taking an expecation with respect to the code distribution, we canthen analyze this error.
Mark M. Wilde (LSU) 94 / 113
![Page 381: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/381.jpg)
Error Analysis
The expected channel output with respect to the code distribution isσB =
∑x pX (x)σxB , which has a typical projection Πσ.
The error probability will ultimately change just slightly byincorporating this projection into the analysis:
Tr{Πσσxn(m)Bn Πσ}−Tr{ΠmΠ̂m−1 · · · Π̂1Πσσ
xn(m)Bn ΠσΠ̂1 · · · Π̂m−1Πm}.
Using a quantum version of the union bound, this can be boundedfrom above by
2
√√√√Tr{(I − Πm)Πσσxn(m)Bn Πσ}+
m−1∑i=1
Tr{ΠiΠσσxn(m)Bn Πσ}
The two terms above are exactly analogous to similar error terms thatarise in the analysis of Shannon’s channel coding theorem.
By taking an expecation with respect to the code distribution, we canthen analyze this error.
Mark M. Wilde (LSU) 94 / 113
![Page 382: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/382.jpg)
Error to bound:
2
√√√√EC{Tr{(I − Πm)ΠσσX n(m)Bn Πσ}}+
m−1∑i=1
EC{Tr{ΠiΠσσX n(m)Bn Πσ}}
The first term can be made small using properties of typicality.
The second term can be made small by choosing the code rate to besmaller than the mutual information I (X ;B) = H(B)− H(B|X ).Consider that
EC{Tr{ΠiΠσσxn(m)Bn Πσ}} = Tr{EX n(i){Πi}ΠσEX n(m){σ
X n(m)Bn }Πσ}
= Tr{EX n(i){Πi}Πσσ⊗nΠσ}
≤ 2−n[H(B)−δ] Tr{EX n(i){Πi}Πσ}}
≤ 2−n[H(B)−δ]EX n(i){Tr{Πi}}
≤ 2−n[H(B)−δ]2n[H(B|X )+δ]
= 2−n[I (X ;B)−2δ].
Mark M. Wilde (LSU) 95 / 113
![Page 383: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/383.jpg)
Error to bound:
2
√√√√EC{Tr{(I − Πm)ΠσσX n(m)Bn Πσ}}+
m−1∑i=1
EC{Tr{ΠiΠσσX n(m)Bn Πσ}}
The first term can be made small using properties of typicality.
The second term can be made small by choosing the code rate to besmaller than the mutual information I (X ;B) = H(B)− H(B|X ).Consider that
EC{Tr{ΠiΠσσxn(m)Bn Πσ}} = Tr{EX n(i){Πi}ΠσEX n(m){σ
X n(m)Bn }Πσ}
= Tr{EX n(i){Πi}Πσσ⊗nΠσ}
≤ 2−n[H(B)−δ] Tr{EX n(i){Πi}Πσ}}
≤ 2−n[H(B)−δ]EX n(i){Tr{Πi}}
≤ 2−n[H(B)−δ]2n[H(B|X )+δ]
= 2−n[I (X ;B)−2δ].
Mark M. Wilde (LSU) 95 / 113
![Page 384: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/384.jpg)
Error to bound:
2
√√√√EC{Tr{(I − Πm)ΠσσX n(m)Bn Πσ}}+
m−1∑i=1
EC{Tr{ΠiΠσσX n(m)Bn Πσ}}
The first term can be made small using properties of typicality.
The second term can be made small by choosing the code rate to besmaller than the mutual information I (X ;B) = H(B)− H(B|X ).Consider that
EC{Tr{ΠiΠσσxn(m)Bn Πσ}} = Tr{EX n(i){Πi}ΠσEX n(m){σ
X n(m)Bn }Πσ}
= Tr{EX n(i){Πi}Πσσ⊗nΠσ}
≤ 2−n[H(B)−δ] Tr{EX n(i){Πi}Πσ}}
≤ 2−n[H(B)−δ]EX n(i){Tr{Πi}}
≤ 2−n[H(B)−δ]2n[H(B|X )+δ]
= 2−n[I (X ;B)−2δ].
Mark M. Wilde (LSU) 95 / 113
![Page 385: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/385.jpg)
Error to bound:
2
√√√√EC{Tr{(I − Πm)ΠσσX n(m)Bn Πσ}}+
m−1∑i=1
EC{Tr{ΠiΠσσX n(m)Bn Πσ}}
The first term can be made small using properties of typicality.
The second term can be made small by choosing the code rate to besmaller than the mutual information I (X ;B) = H(B)− H(B|X ).Consider that
EC{Tr{ΠiΠσσxn(m)Bn Πσ}} = Tr{EX n(i){Πi}ΠσEX n(m){σ
X n(m)Bn }Πσ}
= Tr{EX n(i){Πi}Πσσ⊗nΠσ}
≤ 2−n[H(B)−δ] Tr{EX n(i){Πi}Πσ}}
≤ 2−n[H(B)−δ]EX n(i){Tr{Πi}}
≤ 2−n[H(B)−δ]2n[H(B|X )+δ]
= 2−n[I (X ;B)−2δ].
Mark M. Wilde (LSU) 95 / 113
![Page 386: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/386.jpg)
Conclusion of achievability part
As long as we pick dim(HM) = 2n[I (X ;B)−3δ], then there exists a codewith small error probability, which we can make approach zero bypicking n larger and larger.
We can then expurgate the code if we wish to go from average tomaximal error probability (throw away the worse half of thecodewords, as in the classical case).
So the Holevo information I (X ;B) is an achievable rate.
Mark M. Wilde (LSU) 96 / 113
![Page 387: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/387.jpg)
Conclusion of achievability part
As long as we pick dim(HM) = 2n[I (X ;B)−3δ], then there exists a codewith small error probability, which we can make approach zero bypicking n larger and larger.
We can then expurgate the code if we wish to go from average tomaximal error probability (throw away the worse half of thecodewords, as in the classical case).
So the Holevo information I (X ;B) is an achievable rate.
Mark M. Wilde (LSU) 96 / 113
![Page 388: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/388.jpg)
Conclusion of achievability part
As long as we pick dim(HM) = 2n[I (X ;B)−3δ], then there exists a codewith small error probability, which we can make approach zero bypicking n larger and larger.
We can then expurgate the code if we wish to go from average tomaximal error probability (throw away the worse half of thecodewords, as in the classical case).
So the Holevo information I (X ;B) is an achievable rate.
Mark M. Wilde (LSU) 96 / 113
![Page 389: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/389.jpg)
Conclusion of achievability part
As long as we pick dim(HM) = 2n[I (X ;B)−3δ], then there exists a codewith small error probability, which we can make approach zero bypicking n larger and larger.
We can then expurgate the code if we wish to go from average tomaximal error probability (throw away the worse half of thecodewords, as in the classical case).
So the Holevo information I (X ;B) is an achievable rate.
Mark M. Wilde (LSU) 96 / 113
![Page 390: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/390.jpg)
Converse theorem
The converse part of the theorem establishes the regularized Holevoinformation as an upper bound on classical capacity:
C (N ) ≤ limn→∞
1
nχ(N⊗n).
For some channels, such as entanglement-breaking channels, thefollowing collapse happens for all n:
1
nχ(N⊗n) = χ(N ).
But we know it does not happen in general. That is, it is known thatthere exists a channel for which
χ(N ) < limn→∞
1
nχ(N⊗n).
So there still remains quite a bit to understand about classicalcapacity.
Mark M. Wilde (LSU) 97 / 113
![Page 391: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/391.jpg)
Converse theorem
The converse part of the theorem establishes the regularized Holevoinformation as an upper bound on classical capacity:
C (N ) ≤ limn→∞
1
nχ(N⊗n).
For some channels, such as entanglement-breaking channels, thefollowing collapse happens for all n:
1
nχ(N⊗n) = χ(N ).
But we know it does not happen in general. That is, it is known thatthere exists a channel for which
χ(N ) < limn→∞
1
nχ(N⊗n).
So there still remains quite a bit to understand about classicalcapacity.
Mark M. Wilde (LSU) 97 / 113
![Page 392: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/392.jpg)
Converse theorem
The converse part of the theorem establishes the regularized Holevoinformation as an upper bound on classical capacity:
C (N ) ≤ limn→∞
1
nχ(N⊗n).
For some channels, such as entanglement-breaking channels, thefollowing collapse happens for all n:
1
nχ(N⊗n) = χ(N ).
But we know it does not happen in general. That is, it is known thatthere exists a channel for which
χ(N ) < limn→∞
1
nχ(N⊗n).
So there still remains quite a bit to understand about classicalcapacity.
Mark M. Wilde (LSU) 97 / 113
![Page 393: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/393.jpg)
Converse theorem
The converse part of the theorem establishes the regularized Holevoinformation as an upper bound on classical capacity:
C (N ) ≤ limn→∞
1
nχ(N⊗n).
For some channels, such as entanglement-breaking channels, thefollowing collapse happens for all n:
1
nχ(N⊗n) = χ(N ).
But we know it does not happen in general. That is, it is known thatthere exists a channel for which
χ(N ) < limn→∞
1
nχ(N⊗n).
So there still remains quite a bit to understand about classicalcapacity.
Mark M. Wilde (LSU) 97 / 113
![Page 394: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/394.jpg)
Converse theorem
The converse part of the theorem establishes the regularized Holevoinformation as an upper bound on classical capacity:
C (N ) ≤ limn→∞
1
nχ(N⊗n).
For some channels, such as entanglement-breaking channels, thefollowing collapse happens for all n:
1
nχ(N⊗n) = χ(N ).
But we know it does not happen in general. That is, it is known thatthere exists a channel for which
χ(N ) < limn→∞
1
nχ(N⊗n).
So there still remains quite a bit to understand about classicalcapacity.
Mark M. Wilde (LSU) 97 / 113
![Page 395: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/395.jpg)
Entanglement-assisted comm.
Mark M. Wilde (LSU) 98 / 113
![Page 396: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/396.jpg)
Entanglement-assisted classical communication code
Now allow for Alice and Bob to share entanglement beforecommunication begins. From super-dense coding, we know thatentanglement can double the classical capacity of a noiseless qubitchannel. What about in general?
An (n,R, ε) entanglement-assisted classical comm. code consists ofan encoding channel EM′TA→An , a decoding measurement channelDBnTB→M̂ , and an entangled state ΨTATB
such that:
F (ΦMM̂ , (DBnTB→M̂ ◦ N⊗nA→B ◦ EM′TA→An)(ΦMM′ ⊗ΨTATB
)) ≥ 1− ε,
where
ΦMM̂ ≡1
dim(HM)
∑m
|m〉〈m|M ⊗ |m〉〈m|M̂ ,
and 1n log2(dim(HM)) ≥ R.
The goal again is for the coding scheme to preserve the classicalcorrelations in the state ΦMM̂ .
Mark M. Wilde (LSU) 99 / 113
![Page 397: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/397.jpg)
Entanglement-assisted classical communication code
Now allow for Alice and Bob to share entanglement beforecommunication begins. From super-dense coding, we know thatentanglement can double the classical capacity of a noiseless qubitchannel. What about in general?
An (n,R, ε) entanglement-assisted classical comm. code consists ofan encoding channel EM′TA→An , a decoding measurement channelDBnTB→M̂ , and an entangled state ΨTATB
such that:
F (ΦMM̂ , (DBnTB→M̂ ◦ N⊗nA→B ◦ EM′TA→An)(ΦMM′ ⊗ΨTATB
)) ≥ 1− ε,
where
ΦMM̂ ≡1
dim(HM)
∑m
|m〉〈m|M ⊗ |m〉〈m|M̂ ,
and 1n log2(dim(HM)) ≥ R.
The goal again is for the coding scheme to preserve the classicalcorrelations in the state ΦMM̂ .
Mark M. Wilde (LSU) 99 / 113
![Page 398: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/398.jpg)
Entanglement-assisted classical communication code
Now allow for Alice and Bob to share entanglement beforecommunication begins. From super-dense coding, we know thatentanglement can double the classical capacity of a noiseless qubitchannel. What about in general?
An (n,R, ε) entanglement-assisted classical comm. code consists ofan encoding channel EM′TA→An , a decoding measurement channelDBnTB→M̂ , and an entangled state ΨTATB
such that:
F (ΦMM̂ , (DBnTB→M̂ ◦ N⊗nA→B ◦ EM′TA→An)(ΦMM′ ⊗ΨTATB
)) ≥ 1− ε,
where
ΦMM̂ ≡1
dim(HM)
∑m
|m〉〈m|M ⊗ |m〉〈m|M̂ ,
and 1n log2(dim(HM)) ≥ R.
The goal again is for the coding scheme to preserve the classicalcorrelations in the state ΦMM̂ .
Mark M. Wilde (LSU) 99 / 113
![Page 399: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/399.jpg)
Entanglement-assisted classical communication code
Now allow for Alice and Bob to share entanglement beforecommunication begins. From super-dense coding, we know thatentanglement can double the classical capacity of a noiseless qubitchannel. What about in general?
An (n,R, ε) entanglement-assisted classical comm. code consists ofan encoding channel EM′TA→An , a decoding measurement channelDBnTB→M̂ , and an entangled state ΨTATB
such that:
F (ΦMM̂ , (DBnTB→M̂ ◦ N⊗nA→B ◦ EM′TA→An)(ΦMM′ ⊗ΨTATB
)) ≥ 1− ε,
where
ΦMM̂ ≡1
dim(HM)
∑m
|m〉〈m|M ⊗ |m〉〈m|M̂ ,
and 1n log2(dim(HM)) ≥ R.
The goal again is for the coding scheme to preserve the classicalcorrelations in the state ΦMM̂ .
Mark M. Wilde (LSU) 99 / 113
![Page 400: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/400.jpg)
Entanglement-assisted classical communication code
Now allow for Alice and Bob to share entanglement beforecommunication begins. From super-dense coding, we know thatentanglement can double the classical capacity of a noiseless qubitchannel. What about in general?
An (n,R, ε) entanglement-assisted classical comm. code consists ofan encoding channel EM′TA→An , a decoding measurement channelDBnTB→M̂ , and an entangled state ΨTATB
such that:
F (ΦMM̂ , (DBnTB→M̂ ◦ N⊗nA→B ◦ EM′TA→An)(ΦMM′ ⊗ΨTATB
)) ≥ 1− ε,
where
ΦMM̂ ≡1
dim(HM)
∑m
|m〉〈m|M ⊗ |m〉〈m|M̂ ,
and 1n log2(dim(HM)) ≥ R.
The goal again is for the coding scheme to preserve the classicalcorrelations in the state ΦMM̂ .
Mark M. Wilde (LSU) 99 / 113
![Page 401: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/401.jpg)
Schematic of an EA classical communication code
Mark M. Wilde (LSU) 100 / 113
![Page 402: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/402.jpg)
Entanglement-assisted classical capacity
A rate R for entanglement-assisted (EA) classical communication isachievable if for all ε ∈ (0, 1) and sufficiently large n, there exists an(n,R, ε) EA classical communication code.
The EA classical capacity CEA(N ) of a quantum channel N is equalto the supremum of all achievable rates.
Mark M. Wilde (LSU) 101 / 113
![Page 403: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/403.jpg)
Entanglement-assisted classical capacity
A rate R for entanglement-assisted (EA) classical communication isachievable if for all ε ∈ (0, 1) and sufficiently large n, there exists an(n,R, ε) EA classical communication code.
The EA classical capacity CEA(N ) of a quantum channel N is equalto the supremum of all achievable rates.
Mark M. Wilde (LSU) 101 / 113
![Page 404: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/404.jpg)
Entanglement-assisted classical capacity
A rate R for entanglement-assisted (EA) classical communication isachievable if for all ε ∈ (0, 1) and sufficiently large n, there exists an(n,R, ε) EA classical communication code.
The EA classical capacity CEA(N ) of a quantum channel N is equalto the supremum of all achievable rates.
Mark M. Wilde (LSU) 101 / 113
![Page 405: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/405.jpg)
What is known about entanglement-assisted capacity
Entanglement-assisted capacity theorem:
CEA(N ) = I (N )
where I (N ) = maxφRA
I (R;B)ω,
ωRB ≡ NA→B(φRA).
Thus, this problem is completely solved!
CEA(N ) does not change if there is a quantum feedback channel fromBob to Alice. We even know strong converse theorems for this settingas well. In these senses, the entanglement-assisted capacity representsthe fully quantum analog of Shannon’s channel capacity theorem.
Mark M. Wilde (LSU) 102 / 113
![Page 406: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/406.jpg)
What is known about entanglement-assisted capacity
Entanglement-assisted capacity theorem:
CEA(N ) = I (N )
where I (N ) = maxφRA
I (R;B)ω,
ωRB ≡ NA→B(φRA).
Thus, this problem is completely solved!
CEA(N ) does not change if there is a quantum feedback channel fromBob to Alice. We even know strong converse theorems for this settingas well. In these senses, the entanglement-assisted capacity representsthe fully quantum analog of Shannon’s channel capacity theorem.
Mark M. Wilde (LSU) 102 / 113
![Page 407: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/407.jpg)
What is known about entanglement-assisted capacity
Entanglement-assisted capacity theorem:
CEA(N ) = I (N )
where I (N ) = maxφRA
I (R;B)ω,
ωRB ≡ NA→B(φRA).
Thus, this problem is completely solved!
CEA(N ) does not change if there is a quantum feedback channel fromBob to Alice. We even know strong converse theorems for this settingas well. In these senses, the entanglement-assisted capacity representsthe fully quantum analog of Shannon’s channel capacity theorem.
Mark M. Wilde (LSU) 102 / 113
![Page 408: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/408.jpg)
What is known about entanglement-assisted capacity
Entanglement-assisted capacity theorem:
CEA(N ) = I (N )
where I (N ) = maxφRA
I (R;B)ω,
ωRB ≡ NA→B(φRA).
Thus, this problem is completely solved!
CEA(N ) does not change if there is a quantum feedback channel fromBob to Alice. We even know strong converse theorems for this settingas well. In these senses, the entanglement-assisted capacity representsthe fully quantum analog of Shannon’s channel capacity theorem.
Mark M. Wilde (LSU) 102 / 113
![Page 409: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/409.jpg)
Entanglement-assisted coding (simple version)
Allow Alice and Bob to share a maximally entangled state |Φ〉〈Φ|AB .
They then induce the following ensemble by Alice applying arandomly selected, generalized Pauli operator to her input:{
d−2, (NA→B′ ⊗ idB) (|Φx ,zAB〉〈Φ
x ,zAB |)
}.
where |Φx ,z〉AB = X (x)A Z (z)A |Φ〉AB . (This is the same ensemblefrom super-dense coding if N is the identity channel.)
By previous achievability result and some entropy manipulations, wecan conclude that the mutual information I (B ′;B)N (Φ) is achievable.
More general argument establishes that I (B ′;B)N (φ) is achievable,where φAB is a pure bipartite state. So then CEA(N ) ≥ I (N ).
Mark M. Wilde (LSU) 103 / 113
![Page 410: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/410.jpg)
Entanglement-assisted coding (simple version)
Allow Alice and Bob to share a maximally entangled state |Φ〉〈Φ|AB .
They then induce the following ensemble by Alice applying arandomly selected, generalized Pauli operator to her input:{
d−2, (NA→B′ ⊗ idB) (|Φx ,zAB〉〈Φ
x ,zAB |)
}.
where |Φx ,z〉AB = X (x)A Z (z)A |Φ〉AB . (This is the same ensemblefrom super-dense coding if N is the identity channel.)
By previous achievability result and some entropy manipulations, wecan conclude that the mutual information I (B ′;B)N (Φ) is achievable.
More general argument establishes that I (B ′;B)N (φ) is achievable,where φAB is a pure bipartite state. So then CEA(N ) ≥ I (N ).
Mark M. Wilde (LSU) 103 / 113
![Page 411: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/411.jpg)
Entanglement-assisted coding (simple version)
Allow Alice and Bob to share a maximally entangled state |Φ〉〈Φ|AB .
They then induce the following ensemble by Alice applying arandomly selected, generalized Pauli operator to her input:{
d−2, (NA→B′ ⊗ idB) (|Φx ,zAB〉〈Φ
x ,zAB |)
}.
where |Φx ,z〉AB = X (x)A Z (z)A |Φ〉AB . (This is the same ensemblefrom super-dense coding if N is the identity channel.)
By previous achievability result and some entropy manipulations, wecan conclude that the mutual information I (B ′;B)N (Φ) is achievable.
More general argument establishes that I (B ′;B)N (φ) is achievable,where φAB is a pure bipartite state. So then CEA(N ) ≥ I (N ).
Mark M. Wilde (LSU) 103 / 113
![Page 412: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/412.jpg)
Entanglement-assisted coding (simple version)
Allow Alice and Bob to share a maximally entangled state |Φ〉〈Φ|AB .
They then induce the following ensemble by Alice applying arandomly selected, generalized Pauli operator to her input:{
d−2, (NA→B′ ⊗ idB) (|Φx ,zAB〉〈Φ
x ,zAB |)
}.
where |Φx ,z〉AB = X (x)A Z (z)A |Φ〉AB . (This is the same ensemblefrom super-dense coding if N is the identity channel.)
By previous achievability result and some entropy manipulations, wecan conclude that the mutual information I (B ′;B)N (Φ) is achievable.
More general argument establishes that I (B ′;B)N (φ) is achievable,where φAB is a pure bipartite state. So then CEA(N ) ≥ I (N ).
Mark M. Wilde (LSU) 103 / 113
![Page 413: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/413.jpg)
Entanglement-assisted coding (simple version)
Allow Alice and Bob to share a maximally entangled state |Φ〉〈Φ|AB .
They then induce the following ensemble by Alice applying arandomly selected, generalized Pauli operator to her input:{
d−2, (NA→B′ ⊗ idB) (|Φx ,zAB〉〈Φ
x ,zAB |)
}.
where |Φx ,z〉AB = X (x)A Z (z)A |Φ〉AB . (This is the same ensemblefrom super-dense coding if N is the identity channel.)
By previous achievability result and some entropy manipulations, wecan conclude that the mutual information I (B ′;B)N (Φ) is achievable.
More general argument establishes that I (B ′;B)N (φ) is achievable,where φAB is a pure bipartite state. So then CEA(N ) ≥ I (N ).
Mark M. Wilde (LSU) 103 / 113
![Page 414: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/414.jpg)
Entanglement-assisted converse theorem
Employ data processing and the chain rule for conditional mutualinformation to conclude that
CEA(N ) ≤ I (N ).
Can even establish this bound when there is a quantum feedbackchannel of unlimited dimension connecting Bob to Alice, a setup like
Mark M. Wilde (LSU) 104 / 113
![Page 415: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/415.jpg)
Entanglement-assisted converse theorem
Employ data processing and the chain rule for conditional mutualinformation to conclude that
CEA(N ) ≤ I (N ).
Can even establish this bound when there is a quantum feedbackchannel of unlimited dimension connecting Bob to Alice, a setup like
Mark M. Wilde (LSU) 104 / 113
![Page 416: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/416.jpg)
Entanglement-assisted converse theorem
Employ data processing and the chain rule for conditional mutualinformation to conclude that
CEA(N ) ≤ I (N ).
Can even establish this bound when there is a quantum feedbackchannel of unlimited dimension connecting Bob to Alice, a setup like
Mark M. Wilde (LSU) 104 / 113
![Page 417: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/417.jpg)
Quantum communication
Mark M. Wilde (LSU) 105 / 113
![Page 418: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/418.jpg)
Quantum communication code
Now Alice would like to transmit quantum information intact to orgenerate entanglement with Bob, perhaps for some distributedquantum computation.
An (n,R, ε) quantum communication code consists of an encodingchannel EM′→An and a decoding channel DBn→M̂ such that:
F (ΦMM̂ , (DBn→M̂ ◦ N⊗nA→B ◦ EM′→An)(ΦMM′)) ≥ 1− ε,
where ΦMM̂ is the maximally entangled state:
ΦMM̂ ≡1
dim(HM)
∑m,m′
|m〉〈m′|M ⊗ |m〉〈m′|M̂ ,
and 1n log2(dim(HM)) ≥ R.
The goal now is for the coding scheme to preserve the quantumcorrelations in the state ΦMM̂ .
Mark M. Wilde (LSU) 106 / 113
![Page 419: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/419.jpg)
Quantum communication code
Now Alice would like to transmit quantum information intact to orgenerate entanglement with Bob, perhaps for some distributedquantum computation.
An (n,R, ε) quantum communication code consists of an encodingchannel EM′→An and a decoding channel DBn→M̂ such that:
F (ΦMM̂ , (DBn→M̂ ◦ N⊗nA→B ◦ EM′→An)(ΦMM′)) ≥ 1− ε,
where ΦMM̂ is the maximally entangled state:
ΦMM̂ ≡1
dim(HM)
∑m,m′
|m〉〈m′|M ⊗ |m〉〈m′|M̂ ,
and 1n log2(dim(HM)) ≥ R.
The goal now is for the coding scheme to preserve the quantumcorrelations in the state ΦMM̂ .
Mark M. Wilde (LSU) 106 / 113
![Page 420: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/420.jpg)
Quantum communication code
Now Alice would like to transmit quantum information intact to orgenerate entanglement with Bob, perhaps for some distributedquantum computation.
An (n,R, ε) quantum communication code consists of an encodingchannel EM′→An and a decoding channel DBn→M̂ such that:
F (ΦMM̂ , (DBn→M̂ ◦ N⊗nA→B ◦ EM′→An)(ΦMM′)) ≥ 1− ε,
where ΦMM̂ is the maximally entangled state:
ΦMM̂ ≡1
dim(HM)
∑m,m′
|m〉〈m′|M ⊗ |m〉〈m′|M̂ ,
and 1n log2(dim(HM)) ≥ R.
The goal now is for the coding scheme to preserve the quantumcorrelations in the state ΦMM̂ .
Mark M. Wilde (LSU) 106 / 113
![Page 421: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/421.jpg)
Quantum communication code
Now Alice would like to transmit quantum information intact to orgenerate entanglement with Bob, perhaps for some distributedquantum computation.
An (n,R, ε) quantum communication code consists of an encodingchannel EM′→An and a decoding channel DBn→M̂ such that:
F (ΦMM̂ , (DBn→M̂ ◦ N⊗nA→B ◦ EM′→An)(ΦMM′)) ≥ 1− ε,
where ΦMM̂ is the maximally entangled state:
ΦMM̂ ≡1
dim(HM)
∑m,m′
|m〉〈m′|M ⊗ |m〉〈m′|M̂ ,
and 1n log2(dim(HM)) ≥ R.
The goal now is for the coding scheme to preserve the quantumcorrelations in the state ΦMM̂ .
Mark M. Wilde (LSU) 106 / 113
![Page 422: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/422.jpg)
Quantum communication code
Now Alice would like to transmit quantum information intact to orgenerate entanglement with Bob, perhaps for some distributedquantum computation.
An (n,R, ε) quantum communication code consists of an encodingchannel EM′→An and a decoding channel DBn→M̂ such that:
F (ΦMM̂ , (DBn→M̂ ◦ N⊗nA→B ◦ EM′→An)(ΦMM′)) ≥ 1− ε,
where ΦMM̂ is the maximally entangled state:
ΦMM̂ ≡1
dim(HM)
∑m,m′
|m〉〈m′|M ⊗ |m〉〈m′|M̂ ,
and 1n log2(dim(HM)) ≥ R.
The goal now is for the coding scheme to preserve the quantumcorrelations in the state ΦMM̂ .
Mark M. Wilde (LSU) 106 / 113
![Page 423: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/423.jpg)
Quantum capacity
A rate R for quantum communication is achievable if for all ε ∈ (0, 1)and sufficiently large n, there exists an (n,R, ε) quantumcommunication code.
The quantum capacity Q(N ) of a quantum channel N is equal to thesupremum of all achievable rates.
Mark M. Wilde (LSU) 107 / 113
![Page 424: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/424.jpg)
Quantum capacity
A rate R for quantum communication is achievable if for all ε ∈ (0, 1)and sufficiently large n, there exists an (n,R, ε) quantumcommunication code.
The quantum capacity Q(N ) of a quantum channel N is equal to thesupremum of all achievable rates.
Mark M. Wilde (LSU) 107 / 113
![Page 425: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/425.jpg)
Quantum capacity
A rate R for quantum communication is achievable if for all ε ∈ (0, 1)and sufficiently large n, there exists an (n,R, ε) quantumcommunication code.
The quantum capacity Q(N ) of a quantum channel N is equal to thesupremum of all achievable rates.
Mark M. Wilde (LSU) 107 / 113
![Page 426: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/426.jpg)
What is known about quantum capacity
Coherent information lower bound on quantum capacity:
Ic(N ) ≤ Q(N )
where Ic(N ) = maxφRA
I (R〉B)ω,
ωRB ≡ NA→B(φRA).
If a quantum channel is degradable (meaning that the receiver cansimulate the channel from the input to the environment), then
Ic(N ) = Q(N ).
A number of interesting quantum channels have this property.
Quantum capacity is not known for most non-degradable channels. Italso exhibits a striking effect called superactivation: there existzero-quantum capacity channels such that they can combine to havea non-zero quantum capacity. (This does not occur for the basicsetups in classical information theory.)
Mark M. Wilde (LSU) 108 / 113
![Page 427: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/427.jpg)
What is known about quantum capacity
Coherent information lower bound on quantum capacity:
Ic(N ) ≤ Q(N )
where Ic(N ) = maxφRA
I (R〉B)ω,
ωRB ≡ NA→B(φRA).
If a quantum channel is degradable (meaning that the receiver cansimulate the channel from the input to the environment), then
Ic(N ) = Q(N ).
A number of interesting quantum channels have this property.
Quantum capacity is not known for most non-degradable channels. Italso exhibits a striking effect called superactivation: there existzero-quantum capacity channels such that they can combine to havea non-zero quantum capacity. (This does not occur for the basicsetups in classical information theory.)
Mark M. Wilde (LSU) 108 / 113
![Page 428: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/428.jpg)
What is known about quantum capacity
Coherent information lower bound on quantum capacity:
Ic(N ) ≤ Q(N )
where Ic(N ) = maxφRA
I (R〉B)ω,
ωRB ≡ NA→B(φRA).
If a quantum channel is degradable (meaning that the receiver cansimulate the channel from the input to the environment), then
Ic(N ) = Q(N ).
A number of interesting quantum channels have this property.
Quantum capacity is not known for most non-degradable channels. Italso exhibits a striking effect called superactivation: there existzero-quantum capacity channels such that they can combine to havea non-zero quantum capacity. (This does not occur for the basicsetups in classical information theory.)
Mark M. Wilde (LSU) 108 / 113
![Page 429: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/429.jpg)
What is known about quantum capacity
Coherent information lower bound on quantum capacity:
Ic(N ) ≤ Q(N )
where Ic(N ) = maxφRA
I (R〉B)ω,
ωRB ≡ NA→B(φRA).
If a quantum channel is degradable (meaning that the receiver cansimulate the channel from the input to the environment), then
Ic(N ) = Q(N ).
A number of interesting quantum channels have this property.
Quantum capacity is not known for most non-degradable channels. Italso exhibits a striking effect called superactivation: there existzero-quantum capacity channels such that they can combine to havea non-zero quantum capacity. (This does not occur for the basicsetups in classical information theory.)
Mark M. Wilde (LSU) 108 / 113
![Page 430: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/430.jpg)
Achieving the coherent information
There are now many coding methods known for achieving thecoherent information rate.
Perhaps the most prominent is known as the decoupling method.
Suppose that Alice, Bob, and Eve share a tripartite pure entangledstate |ψ〉〈ψ|RBE after Alice transmits her share of the entanglementwith the reference through a noisy channel.
Then if the reduced state ψRE on the reference system and Eve’ssystem is approximately decoupled, meaning that
‖ψRE − ψR ⊗ σE‖1 ≤ ε,
where σE is arbitrary state, this implies that Bob can decode quantuminformation that Alice intended to send to him. Can show thatdecoupling is possible as long as qubit rate ≈ coherent information.
Mark M. Wilde (LSU) 109 / 113
![Page 431: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/431.jpg)
Achieving the coherent information
There are now many coding methods known for achieving thecoherent information rate.
Perhaps the most prominent is known as the decoupling method.
Suppose that Alice, Bob, and Eve share a tripartite pure entangledstate |ψ〉〈ψ|RBE after Alice transmits her share of the entanglementwith the reference through a noisy channel.
Then if the reduced state ψRE on the reference system and Eve’ssystem is approximately decoupled, meaning that
‖ψRE − ψR ⊗ σE‖1 ≤ ε,
where σE is arbitrary state, this implies that Bob can decode quantuminformation that Alice intended to send to him. Can show thatdecoupling is possible as long as qubit rate ≈ coherent information.
Mark M. Wilde (LSU) 109 / 113
![Page 432: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/432.jpg)
Achieving the coherent information
There are now many coding methods known for achieving thecoherent information rate.
Perhaps the most prominent is known as the decoupling method.
Suppose that Alice, Bob, and Eve share a tripartite pure entangledstate |ψ〉〈ψ|RBE after Alice transmits her share of the entanglementwith the reference through a noisy channel.
Then if the reduced state ψRE on the reference system and Eve’ssystem is approximately decoupled, meaning that
‖ψRE − ψR ⊗ σE‖1 ≤ ε,
where σE is arbitrary state, this implies that Bob can decode quantuminformation that Alice intended to send to him. Can show thatdecoupling is possible as long as qubit rate ≈ coherent information.
Mark M. Wilde (LSU) 109 / 113
![Page 433: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/433.jpg)
Achieving the coherent information
There are now many coding methods known for achieving thecoherent information rate.
Perhaps the most prominent is known as the decoupling method.
Suppose that Alice, Bob, and Eve share a tripartite pure entangledstate |ψ〉〈ψ|RBE after Alice transmits her share of the entanglementwith the reference through a noisy channel.
Then if the reduced state ψRE on the reference system and Eve’ssystem is approximately decoupled, meaning that
‖ψRE − ψR ⊗ σE‖1 ≤ ε,
where σE is arbitrary state, this implies that Bob can decode quantuminformation that Alice intended to send to him. Can show thatdecoupling is possible as long as qubit rate ≈ coherent information.
Mark M. Wilde (LSU) 109 / 113
![Page 434: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/434.jpg)
Achieving the coherent information
There are now many coding methods known for achieving thecoherent information rate.
Perhaps the most prominent is known as the decoupling method.
Suppose that Alice, Bob, and Eve share a tripartite pure entangledstate |ψ〉〈ψ|RBE after Alice transmits her share of the entanglementwith the reference through a noisy channel.
Then if the reduced state ψRE on the reference system and Eve’ssystem is approximately decoupled, meaning that
‖ψRE − ψR ⊗ σE‖1 ≤ ε,
where σE is arbitrary state, this implies that Bob can decode quantuminformation that Alice intended to send to him. Can show thatdecoupling is possible as long as qubit rate ≈ coherent information.
Mark M. Wilde (LSU) 109 / 113
![Page 435: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/435.jpg)
Decoupling method
Why does this work? Suppose the state is exactly decoupled. Thenone purification of the state ψRE is the state |ψ〉〈ψ|RBE that theyshare after the channel acts.
Another purification of ψRE = ψR ⊗ σE is |ψ〉〈ψ|RB1 ⊗ |σ〉〈σ|B2E ,where |ψ〉〈ψ|RB1 is the original state that Alice sent through thechannel and |σ〉〈σ|B2E is some other state that purifies the state σEof the environment.
All purifications are related by isometries and Bob possesses thepurification of R and E ,
⇒ There exists some unitary UB→B1B2 such that
UB→B1B2 |ψ〉RBE = |ψ〉RB1 ⊗ |σ〉B2E.
This unitary is then Bob’s decoder!
Thus, the decoupling condition implies the existence of a decoder forBob, so that it is only necessary to show the existence of an encoderthat decouples the reference from the environment.
Mark M. Wilde (LSU) 110 / 113
![Page 436: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/436.jpg)
Decoupling method
Why does this work? Suppose the state is exactly decoupled. Thenone purification of the state ψRE is the state |ψ〉〈ψ|RBE that theyshare after the channel acts.
Another purification of ψRE = ψR ⊗ σE is |ψ〉〈ψ|RB1 ⊗ |σ〉〈σ|B2E ,where |ψ〉〈ψ|RB1 is the original state that Alice sent through thechannel and |σ〉〈σ|B2E is some other state that purifies the state σEof the environment.
All purifications are related by isometries and Bob possesses thepurification of R and E ,
⇒ There exists some unitary UB→B1B2 such that
UB→B1B2 |ψ〉RBE = |ψ〉RB1 ⊗ |σ〉B2E.
This unitary is then Bob’s decoder!
Thus, the decoupling condition implies the existence of a decoder forBob, so that it is only necessary to show the existence of an encoderthat decouples the reference from the environment.
Mark M. Wilde (LSU) 110 / 113
![Page 437: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/437.jpg)
Decoupling method
Why does this work? Suppose the state is exactly decoupled. Thenone purification of the state ψRE is the state |ψ〉〈ψ|RBE that theyshare after the channel acts.
Another purification of ψRE = ψR ⊗ σE is |ψ〉〈ψ|RB1 ⊗ |σ〉〈σ|B2E ,where |ψ〉〈ψ|RB1 is the original state that Alice sent through thechannel and |σ〉〈σ|B2E is some other state that purifies the state σEof the environment.
All purifications are related by isometries and Bob possesses thepurification of R and E ,
⇒ There exists some unitary UB→B1B2 such that
UB→B1B2 |ψ〉RBE = |ψ〉RB1 ⊗ |σ〉B2E.
This unitary is then Bob’s decoder!
Thus, the decoupling condition implies the existence of a decoder forBob, so that it is only necessary to show the existence of an encoderthat decouples the reference from the environment.
Mark M. Wilde (LSU) 110 / 113
![Page 438: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/438.jpg)
Decoupling method
Why does this work? Suppose the state is exactly decoupled. Thenone purification of the state ψRE is the state |ψ〉〈ψ|RBE that theyshare after the channel acts.
Another purification of ψRE = ψR ⊗ σE is |ψ〉〈ψ|RB1 ⊗ |σ〉〈σ|B2E ,where |ψ〉〈ψ|RB1 is the original state that Alice sent through thechannel and |σ〉〈σ|B2E is some other state that purifies the state σEof the environment.
All purifications are related by isometries and Bob possesses thepurification of R and E ,
⇒ There exists some unitary UB→B1B2 such that
UB→B1B2 |ψ〉RBE = |ψ〉RB1 ⊗ |σ〉B2E.
This unitary is then Bob’s decoder!
Thus, the decoupling condition implies the existence of a decoder forBob, so that it is only necessary to show the existence of an encoderthat decouples the reference from the environment.
Mark M. Wilde (LSU) 110 / 113
![Page 439: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/439.jpg)
Decoupling method
Why does this work? Suppose the state is exactly decoupled. Thenone purification of the state ψRE is the state |ψ〉〈ψ|RBE that theyshare after the channel acts.
Another purification of ψRE = ψR ⊗ σE is |ψ〉〈ψ|RB1 ⊗ |σ〉〈σ|B2E ,where |ψ〉〈ψ|RB1 is the original state that Alice sent through thechannel and |σ〉〈σ|B2E is some other state that purifies the state σEof the environment.
All purifications are related by isometries and Bob possesses thepurification of R and E ,
⇒ There exists some unitary UB→B1B2 such that
UB→B1B2 |ψ〉RBE = |ψ〉RB1 ⊗ |σ〉B2E.
This unitary is then Bob’s decoder!
Thus, the decoupling condition implies the existence of a decoder forBob, so that it is only necessary to show the existence of an encoderthat decouples the reference from the environment.
Mark M. Wilde (LSU) 110 / 113
![Page 440: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/440.jpg)
Decoupling method
Why does this work? Suppose the state is exactly decoupled. Thenone purification of the state ψRE is the state |ψ〉〈ψ|RBE that theyshare after the channel acts.
Another purification of ψRE = ψR ⊗ σE is |ψ〉〈ψ|RB1 ⊗ |σ〉〈σ|B2E ,where |ψ〉〈ψ|RB1 is the original state that Alice sent through thechannel and |σ〉〈σ|B2E is some other state that purifies the state σEof the environment.
All purifications are related by isometries and Bob possesses thepurification of R and E ,
⇒ There exists some unitary UB→B1B2 such that
UB→B1B2 |ψ〉RBE = |ψ〉RB1 ⊗ |σ〉B2E.
This unitary is then Bob’s decoder!
Thus, the decoupling condition implies the existence of a decoder forBob, so that it is only necessary to show the existence of an encoderthat decouples the reference from the environment.
Mark M. Wilde (LSU) 110 / 113
![Page 441: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/441.jpg)
Future directions
Mark M. Wilde (LSU) 111 / 113
![Page 442: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/442.jpg)
Open questions
It might be difficult to find a general formula for quantum capacity.Some suspect that the quantity is uncomputable.
Other capacities: private capacity, locking capacity, data hidingcapacity (some results known but many questions remain).
Constructing codes for quantum channels. Major open question forquantum polar codes is to find an efficiently implementable decoder.
Network quantum information theory: Some results known formultiple access, broadcast, interference, relay channels. Major openquestion is to prove the existence of a quantum simultaneous decoder(special cases known, but general case is open).
Strong converses and 2nd-order asymptotics. Some results known.Major open question to establish strong converse property forquantum capacity of degradable channels. Open: 2nd-orderasymptotics for entanglement-assisted capacity of all channels.
Mark M. Wilde (LSU) 112 / 113
![Page 443: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/443.jpg)
Open questions
It might be difficult to find a general formula for quantum capacity.Some suspect that the quantity is uncomputable.
Other capacities: private capacity, locking capacity, data hidingcapacity (some results known but many questions remain).
Constructing codes for quantum channels. Major open question forquantum polar codes is to find an efficiently implementable decoder.
Network quantum information theory: Some results known formultiple access, broadcast, interference, relay channels. Major openquestion is to prove the existence of a quantum simultaneous decoder(special cases known, but general case is open).
Strong converses and 2nd-order asymptotics. Some results known.Major open question to establish strong converse property forquantum capacity of degradable channels. Open: 2nd-orderasymptotics for entanglement-assisted capacity of all channels.
Mark M. Wilde (LSU) 112 / 113
![Page 444: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/444.jpg)
Open questions
It might be difficult to find a general formula for quantum capacity.Some suspect that the quantity is uncomputable.
Other capacities: private capacity, locking capacity, data hidingcapacity (some results known but many questions remain).
Constructing codes for quantum channels. Major open question forquantum polar codes is to find an efficiently implementable decoder.
Network quantum information theory: Some results known formultiple access, broadcast, interference, relay channels. Major openquestion is to prove the existence of a quantum simultaneous decoder(special cases known, but general case is open).
Strong converses and 2nd-order asymptotics. Some results known.Major open question to establish strong converse property forquantum capacity of degradable channels. Open: 2nd-orderasymptotics for entanglement-assisted capacity of all channels.
Mark M. Wilde (LSU) 112 / 113
![Page 445: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/445.jpg)
Open questions
It might be difficult to find a general formula for quantum capacity.Some suspect that the quantity is uncomputable.
Other capacities: private capacity, locking capacity, data hidingcapacity (some results known but many questions remain).
Constructing codes for quantum channels. Major open question forquantum polar codes is to find an efficiently implementable decoder.
Network quantum information theory: Some results known formultiple access, broadcast, interference, relay channels. Major openquestion is to prove the existence of a quantum simultaneous decoder(special cases known, but general case is open).
Strong converses and 2nd-order asymptotics. Some results known.Major open question to establish strong converse property forquantum capacity of degradable channels. Open: 2nd-orderasymptotics for entanglement-assisted capacity of all channels.
Mark M. Wilde (LSU) 112 / 113
![Page 446: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/446.jpg)
Open questions
It might be difficult to find a general formula for quantum capacity.Some suspect that the quantity is uncomputable.
Other capacities: private capacity, locking capacity, data hidingcapacity (some results known but many questions remain).
Constructing codes for quantum channels. Major open question forquantum polar codes is to find an efficiently implementable decoder.
Network quantum information theory: Some results known formultiple access, broadcast, interference, relay channels. Major openquestion is to prove the existence of a quantum simultaneous decoder(special cases known, but general case is open).
Strong converses and 2nd-order asymptotics. Some results known.Major open question to establish strong converse property forquantum capacity of degradable channels. Open: 2nd-orderasymptotics for entanglement-assisted capacity of all channels.
Mark M. Wilde (LSU) 112 / 113
![Page 447: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/447.jpg)
Open questions
It might be difficult to find a general formula for quantum capacity.Some suspect that the quantity is uncomputable.
Other capacities: private capacity, locking capacity, data hidingcapacity (some results known but many questions remain).
Constructing codes for quantum channels. Major open question forquantum polar codes is to find an efficiently implementable decoder.
Network quantum information theory: Some results known formultiple access, broadcast, interference, relay channels. Major openquestion is to prove the existence of a quantum simultaneous decoder(special cases known, but general case is open).
Strong converses and 2nd-order asymptotics. Some results known.Major open question to establish strong converse property forquantum capacity of degradable channels. Open: 2nd-orderasymptotics for entanglement-assisted capacity of all channels.
Mark M. Wilde (LSU) 112 / 113
![Page 448: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/448.jpg)
Other topics
Capacities of Gaussian quantum channels. These model practicalcommunication channels. A number of open questions remain here.(see Shannon lecture of Holevo).
Covert communication over quantum channels. (Informal workshopon Wednesday afternoon).
Quantum channels with memory.
Security of quantum cryptography (bringing theoretical securityproofs closer to experimental implementations).
Reformulating thermodynamics in the quantum regime using sometools of quantum information theory.
Quantifying entanglement (resource theory of entanglement).
Strengthenings of fundamental quantum entropy inequalities.
Mark M. Wilde (LSU) 113 / 113
![Page 449: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/449.jpg)
Other topics
Capacities of Gaussian quantum channels. These model practicalcommunication channels. A number of open questions remain here.(see Shannon lecture of Holevo).
Covert communication over quantum channels. (Informal workshopon Wednesday afternoon).
Quantum channels with memory.
Security of quantum cryptography (bringing theoretical securityproofs closer to experimental implementations).
Reformulating thermodynamics in the quantum regime using sometools of quantum information theory.
Quantifying entanglement (resource theory of entanglement).
Strengthenings of fundamental quantum entropy inequalities.
Mark M. Wilde (LSU) 113 / 113
![Page 450: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/450.jpg)
Other topics
Capacities of Gaussian quantum channels. These model practicalcommunication channels. A number of open questions remain here.(see Shannon lecture of Holevo).
Covert communication over quantum channels. (Informal workshopon Wednesday afternoon).
Quantum channels with memory.
Security of quantum cryptography (bringing theoretical securityproofs closer to experimental implementations).
Reformulating thermodynamics in the quantum regime using sometools of quantum information theory.
Quantifying entanglement (resource theory of entanglement).
Strengthenings of fundamental quantum entropy inequalities.
Mark M. Wilde (LSU) 113 / 113
![Page 451: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/451.jpg)
Other topics
Capacities of Gaussian quantum channels. These model practicalcommunication channels. A number of open questions remain here.(see Shannon lecture of Holevo).
Covert communication over quantum channels. (Informal workshopon Wednesday afternoon).
Quantum channels with memory.
Security of quantum cryptography (bringing theoretical securityproofs closer to experimental implementations).
Reformulating thermodynamics in the quantum regime using sometools of quantum information theory.
Quantifying entanglement (resource theory of entanglement).
Strengthenings of fundamental quantum entropy inequalities.
Mark M. Wilde (LSU) 113 / 113
![Page 452: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/452.jpg)
Other topics
Capacities of Gaussian quantum channels. These model practicalcommunication channels. A number of open questions remain here.(see Shannon lecture of Holevo).
Covert communication over quantum channels. (Informal workshopon Wednesday afternoon).
Quantum channels with memory.
Security of quantum cryptography (bringing theoretical securityproofs closer to experimental implementations).
Reformulating thermodynamics in the quantum regime using sometools of quantum information theory.
Quantifying entanglement (resource theory of entanglement).
Strengthenings of fundamental quantum entropy inequalities.
Mark M. Wilde (LSU) 113 / 113
![Page 453: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/453.jpg)
Other topics
Capacities of Gaussian quantum channels. These model practicalcommunication channels. A number of open questions remain here.(see Shannon lecture of Holevo).
Covert communication over quantum channels. (Informal workshopon Wednesday afternoon).
Quantum channels with memory.
Security of quantum cryptography (bringing theoretical securityproofs closer to experimental implementations).
Reformulating thermodynamics in the quantum regime using sometools of quantum information theory.
Quantifying entanglement (resource theory of entanglement).
Strengthenings of fundamental quantum entropy inequalities.
Mark M. Wilde (LSU) 113 / 113
![Page 454: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/454.jpg)
Other topics
Capacities of Gaussian quantum channels. These model practicalcommunication channels. A number of open questions remain here.(see Shannon lecture of Holevo).
Covert communication over quantum channels. (Informal workshopon Wednesday afternoon).
Quantum channels with memory.
Security of quantum cryptography (bringing theoretical securityproofs closer to experimental implementations).
Reformulating thermodynamics in the quantum regime using sometools of quantum information theory.
Quantifying entanglement (resource theory of entanglement).
Strengthenings of fundamental quantum entropy inequalities.
Mark M. Wilde (LSU) 113 / 113
![Page 455: Quantum Information Theory Tutorialmarkwilde.com/teaching/2016-isit/ISIT-QIT-tutorial-staggered.pdf · published by Cambridge University Press (2nd edition forthcoming) July 10, 2016,](https://reader035.vdocuments.mx/reader035/viewer/2022071015/5fce04399bf4a33c8558270e/html5/thumbnails/455.jpg)
Other topics
Capacities of Gaussian quantum channels. These model practicalcommunication channels. A number of open questions remain here.(see Shannon lecture of Holevo).
Covert communication over quantum channels. (Informal workshopon Wednesday afternoon).
Quantum channels with memory.
Security of quantum cryptography (bringing theoretical securityproofs closer to experimental implementations).
Reformulating thermodynamics in the quantum regime using sometools of quantum information theory.
Quantifying entanglement (resource theory of entanglement).
Strengthenings of fundamental quantum entropy inequalities.
Mark M. Wilde (LSU) 113 / 113