rigorous fault-tolerance thresholds · stabilizer op. fault-tolerance universal fault-tolerance...
TRANSCRIPT
![Page 1: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/1.jpg)
Rigorous fault-tolerancethresholds
Ben ReichardtUC Berkeley
![Page 2: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/2.jpg)
0/1
N gate circuit
![Page 3: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/3.jpg)
N gate circuit
1/0
Need error 1/N
![Page 4: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/4.jpg)
0/1Fault-tolerant, largerC
High tolerable noise Low overhead
Quantum fault-tolerance problem– Classical fault-tolerance: Von Neumann (1956)
![Page 5: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/5.jpg)
– Classical fault-tolerance: Von Neumann (1956)
0/1
Fault-tolerant, larger
C
High tolerable noise Low overhead
Important problem!
Quantum fault-tolerance problem
![Page 6: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/6.jpg)
EC
EC
Intuition
• Aharonov & Ben-Or (‘97), Kitaev (‘97), Knill-Laflamme-Zurek (‘97)– Using concatenated constant-sized code, tolerate constant error
• Quantum fault-tolerance: Shor (1996)– Using poly(log N)-sized code, tolerate 1/poly(log N) error
Work on encoded data Correct errors to prevent spread Concatenate procedure for arbitrary reliability
![Page 7: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/7.jpg)
Intuition Work on encoded data Correct errors to prevent spread Concatenate procedure for arbitrary reliability
• Aharonov & Ben-Or (‘97), Kitaev (‘97), Knill-Laflamme-Zurek (‘97)– Using concatenated constant-sized code, tolerate constant error
EC
EC
• Quantum fault-tolerance: Shor (1996)– Using poly(log N)-sized code, tolerate 1/poly(log N) error
![Page 8: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/8.jpg)
~p(t+1)2
p(t+1)3
p
c pt+1
Probability of error
1
Physical bits per logical bit
m
m2
m3
O(log log N) concatenationspoly(log N) physical bits / logical
• N gate circuitWant error 1/N
1/c1/t
• m-qubit, t-error correcting code
p
c pt+1
Concatenation0.0050.010.0150.020.0250.0050.010.0150.020.025
Physical gate error rate p
Logi
cal g
ate
erro
r rat
e
Logi
cal g
ate
erro
r rat
e
![Page 9: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/9.jpg)
Stabilizer op.fault-tolerance
Universalfault-tolerance
Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04]
– Universality method, related to best current threshold upper bounds– Reduction
B
![Page 10: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/10.jpg)
Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04]– Universality method, related to best current threshold upper bounds– Reduction from FT universality to FT stabilizer operations
Optimized fault-tolerance schemes: [Knill ‘03]– Erasure error threshold is 1/2 for Bell measurements– [Knill ‘05]: > 5% estimated threshold for depolarizing noise
Recent results
1% with substantial but more reasonable overhead
Fault-tolerance threshold myth: Threshold is all that counts.
Maximize the threshold at all costs.
![Page 11: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/11.jpg)
Logi
cal
oper
atio
ns
X Z
mZ
mX
Phy
sica
lop
erat
ions
data
ancillamZ
mZ
X
X
X
appl
y cor
rect
ion
Steane-typeerror correction
![Page 12: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/12.jpg)
X Z
data
ancilla
mZ
mX
Steane-typeerror correction
Knill-typeerror correction
Teleportation
mZ
mX
mZ
mZap
ply c
orre
ctio
n
X
X
X
data
ancilla
mX
mX
mZ
mZ
X
X
X
Logi
cal
oper
atio
nsP
hysi
cal
oper
atio
ns
![Page 13: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/13.jpg)
Logi
cal
oper
atio
nsP
hysi
cal
oper
atio
ns
Teleportation
mZ
mX
data
ancilla
mX
mX
mZ
mZ
Knill-typeerror correction
Advantages– Efficient– Technical advantage: Reduces blockwiseindependence to encoded Bell state
![Page 14: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/14.jpg)
data
ancilla
mX
mX
mZ
mZ
UL
Logi
cal
oper
atio
nsP
hysi
cal
oper
atio
nsKnill-type correction+ computation
Teleportation
mZ
mX
U
Advantages– Efficient– Technical advantage: Reduces blockwiseindependence to encoded Bell state
![Page 15: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/15.jpg)
Logi
cal
oper
atio
nsKnill-type correction+ computation
Teleportation
mZ
mX
U
Phy
sica
lop
erat
ions
data
ancilla
mX
mX
mZ
mZ
UL
Advantages– Efficient– Technical advantage: Reduces blockwiseindependence to encoded Bell state
![Page 16: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/16.jpg)
Logi
cal
oper
atio
nsKnill-type correction+ computation
Teleportation
mZ
mX
U
Phy
sica
lop
erat
ions
data
ancilla
mX
mX
mZ
mZ
UL
Advantages– Efficient– Technical advantage: Reduces blockwiseindependence to encoded Bell state– Allows for more checking
Disadvantages– High overhead at high error rateswith error detection– Renormalization penalty requiresstronger control over error distribution– No threshold has been proved toexist
+ Distance-two code+ Postselection
![Page 17: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/17.jpg)
… …
w/ prob. 1-q w/ prob. q
Main issues Bounded dependencies
– Between different blocks– In time– Between bit errors and logical errors
Example:
(1-q) .97n q .99naccepted w/ prob.
3% bit error rate 1% bit error rate
⇒ Probability of logical error increases exponentially!
![Page 18: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/18.jpg)
Bounded dependencies– Between bit & logical errors
000111
01 ?0? ?’Monotonicity?
Main issues …w/ prob. 1-q
3% bit error rate
…w/ prob. q
1% bit error rate
want encoded Bell pair:
get:
low biterror rate
high biterror rate
monotonicity ⇒
But!
![Page 19: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/19.jpg)
Bounded dependencies– Between bit & logical errors
Monotonicity?
Main issues …w/ prob. 1-q
3% bit error rate
…w/ prob. q
1% bit error rate
(repetition code)
![Page 20: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/20.jpg)
Recent results (continued)
– εc > 0 for several noise models [Aharonov & Ben-Or ‘97, Kitaev ‘97]– εc ≥ 10-4 for independent probabilistic noise [Ouyang, R. ‘05]– εc = 1/2 for Bell measurement erasure errors (detected errors) [Knill ‘03]
Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04]– Universality method, related to best current threshold upper bounds– Reduction from FT universality to FT stabilizer operations
Optimized fault-tolerance schemes: [Knill ‘03]– Erasure error threshold is 1/2 for Bell measurements– [Knill ‘05]: > 5% estimated threshold for depolarizing noise
Improved threshold proofs– Aliferis/Gottesman/Preskill ‘05: 2.7 x 10-5
– R. ‘05: < 1.4 x 10-5
– Ouyang, R. (unpublished): 10-4
more efficientdistance three
1% with substantial but more reasonable overhead
![Page 21: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/21.jpg)
Distance-3 code thresholds Basic estimates
– Aharonov & Ben-Or (1997)– Knill-Laflamme-Zurek (1998)– Preskill (1998)– Gottesman (1997)
Optimized estimates– Zalka (1997)– R. (2004)– Svore-Cross-Chuang-Aho (2005)
2-dimensional locality constraint– Szkopek et al (2004)– Svore-Terhal-DiVincenzo (2005)
But no constant threshold was even proven to exist for distance-3 codes!– Aharonov & Ben-Or proof only works for codes of distance at least 5
Today: Threshold for distance-3 codes
![Page 22: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/22.jpg)
Dist-2 code threshold & threshold gap Knill (2005) has highest threshold estimate ~5%
– … Albeit with large constant overhead (more reasonable at 1%)– Again, no threshold has been proved to exist
Gaps between proven and estimated thresholds– Estimates are as high as ~5%– Aliferis-Gottesman-Preskill (2005): 2.6 x 10-5
Caveat: Small codes aren’t necessarily the most efficient– Steane (‘03) found 23-qubit Golay code had higher threshold (basedon simulations), particularly with slow measurements– 23-qubit Golay code proven: 10-4
![Page 23: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/23.jpg)
Distance-three code threshold proof intuition
Idea: Maintain inductive invariant of wellness. (A block is well “if it has atmost one unwell subblock, and that only rarely.”)
EC
EC
well
well
well
well
What’s new: Control probability distribution oferrors, not just error states.
![Page 24: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/24.jpg)
Aharonov/Ben-Or-style proof intuition
Idea: Maintain inductive invariant of goodness. (A block is good “if it has atmost one bad subblock.”)
EC
ECX
Xgood
good
good
good
(assuming one level k-1 error, m≥7)
X
X
X
X
X
![Page 25: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/25.jpg)
Aharonov/Ben-Or-style proof intuition
Idea: Maintain inductive invariant of goodness. (A block is good “if it has atmost one bad subblock.”)
EC
ECX
Xgood
good
good
good
(assuming one level k-1 error, m≥7)
X
X
X
X
![Page 26: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/26.jpg)
Aharonov/Ben-Or-style proof intuition
Idea: Maintain inductive invariant of goodness. (A block is good “if it has atmost one bad subblock.”)
EC
ECX
Xgood
good
good
bad
(two level k-1 errors, m=7)
X
X
X
XX
XX
X
X
XXXXX
![Page 27: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/27.jpg)
Aharonov/Ben-Or-style proof intuition
Idea: Maintain inductive invariant of goodness. (A block is good “if it has atmost one bad subblock.”)
EC
ECX
Xgood
good
bad
good
(two level k-1 errors)
X
X
X
XX
level-k CNOTfailure rate
level-(k-1)failure rate
# CNOT locations
![Page 28: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/28.jpg)
Aharonov/Ben-Or-style proof intuition
Idea: Maintain inductive invariant of goodness. (A block is good “if it has atmost one bad subblock.”)
EC
EC
For distance-5 code:
EC
ECX
Xgood
good
good
good
X
X
X
![Page 29: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/29.jpg)
Aharonov/Ben-Or-style proof intuition
Idea: Maintain inductive invariant of goodness. (A block is good “if it has atmost one bad subblock.”)
EC
EC
For distance-5 code:
EC
ECX
Xgood
good
good
good
X
X
X
1.
Inefficient:
2. not (distance = 5)
3. No threshold for concatenateddistance-three codes.
![Page 30: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/30.jpg)
Aharonov/Ben-Or-style proof intuition
Idea: Maintain inductive invariant of goodness. (A block is good “if it has atmost one bad subblock.”) Why not for distance-three codes?
ECX
good badXX
New idea: Most blocks should have no bad subblocks. Maintain inductiveinvariant of a controlled probability distribution of errors: “wellness.” (A block iswell “if it only rarely has a bad subblock.”)
(one level k-1 error is already too many)
![Page 31: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/31.jpg)
Def: Error states Def: Relative error states Def: good block Def: “well” block Distance-3 code threshold setup Def: Logical success and failure Distance-3 code threshold proof
Proof overview
(resolve ambiguity)
(encoded CNOT must work even on erroneous input)
![Page 32: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/32.jpg)
X X
X
X X
a
b
a
a b
Def: CNOT
Def: Error states Problem: Different errors areequivalent, so it is ambiguous whichbit is in error
Solution: Track errors from theirintroduction
![Page 33: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/33.jpg)
Tracking errors
X
X
X
I
X X
X
X X
a
b
a
a b
Def: CNOT
X
X X
X
X X
X
I
X
X
X
Problem: Different errors areequivalent, so it is ambiguous whichbit is in error
Solution: Track errors from theirintroduction
Def: Error states
![Page 34: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/34.jpg)
Def: Error states Tracking errors Block error states: ideal recursive
decoding
Note: Block errors do not follow samerules as bit errors
- e.g., 001 + 010 = 011
X
X
X
IX
X X
X
X X
X
I
X
X
X
![Page 35: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/35.jpg)
Tracking errors Block error states: ideal recursivedecoding
Note: relative state is relative to stateof superblock, not superblock’s relativestate We can measure block relative states.
Def: Relative Error states
X
X
X
IX
X X
X
X X
X
I
X
X
X
Relative error states
![Page 36: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/36.jpg)
Tracking errors Block error states: ideal recursive decoding
Relative error states
rel. Xrel. X
rel. X
X X
XXX
Def: A blockk is goodk if it has at most one subblockk-1 either in relative error ornot goodk-1 itself.(Every bit [≡ block0] is good0.)
Def: good
![Page 37: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/37.jpg)
A good block has atmost one subblock eitherin relative error or bad.
Relative error statesbased on ideal recursivedecoding
good bad
good examples
![Page 38: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/38.jpg)
good examples A good block has atmost one subblock eitherin relative error or bad.
Relative error statesbased on ideal recursivedecoding
good bad
![Page 39: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/39.jpg)
good examples A good block has atmost one subblock eitherin relative error or bad.
Relative error statesbased on ideal recursivedecoding
good bad
![Page 40: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/40.jpg)
(at most one subblock either inrelative error or bad)
good bad
![Page 41: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/41.jpg)
(at most one subblock either inrelative error or bad)
good bad
![Page 42: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/42.jpg)
(at most one subblock either inrelative error or bad)
good bad
![Page 43: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/43.jpg)
Tracking errors Block error states: ideal recursive decoding
Relative error statesX X
XXX
Def: A blockk is goodk if it has at most one subblockk-1 either in relative error ornot goodk-1 itself.(Every bit [≡ block0] is good0.)
Def: well
![Page 44: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/44.jpg)
Def: A blockk is wellk(p1,…,pk) if it has at most one subblockk-1 either in relativeerror or not wellk-1(p1,…,pk-1) itself. Additionally, the probability of such a subblock, conditioned on the block’sstate and the state of all bits in other blocks, is ≤ pk.(Every bit [≡ block0] is well0.)
Tracking errors Block error states: ideal recursive decoding
Relative error statesX X
XXX
Def: well
![Page 45: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/45.jpg)
Def: A blockk is wellk(p1,…,pk) if it has at most one subblockk-1 either in relativeerror or not wellk-1(p1,…,pk-1) itself. Additionally, the probability of such a subblock, conditioned on the block’sstate and the state of all bits in other blocks, is ≤ pk.(Every bit [≡ block0] is well0.)
Def: well Tracking errors Block error states: ideal recursive decoding
Relative error statesX X
XXX
Note: Conditioned on block’s state, e.g.,
is not 1-well.w/prob. 1-pk w/prob. pk
![Page 46: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/46.jpg)
Dist-3 code setup
Claim Ck (CNOTk): On success:– Wellk(b1,…,bk) inputs ⇒ wellk(b1,…,bk) outputs, and logical CNOT– Arbitrary inputs ⇒ wellk(b1,…,bk) outputs, and possibly incorrect logical effect
Failure prob. ≤ Ck (C0 = p).
Base noise model: CNOT0 gates fail with X errors independently w/ prob. p
![Page 47: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/47.jpg)
Def: Logical operation Uk on one or more blocksk has the correct logicaleffect if the diagram commutes:
Uk has a possibly incorrect logical effect if the same diagram commutes butwith on the top arrow, where P is a Pauli operator or Pauli product onthe involved blocks.
Def: Logical failure
![Page 48: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/48.jpg)
Claim Ck (CNOTk): On success:– Well inputs ⇒ well outputs, and logical CNOT– Arbitrary inputs ⇒ well outputs
Failure prob. ≤ Ck (C0 = p).
EC
EC
well
well
D
D
D
D
Claim Bk (Correctionk): On success:– Wellk(b1,…,bk) input ⇒ wellk(b1,…,bk) output, and no logical effect– Arbitrary input ⇒ wellk(b1,…,bk) output
Failure prob. ≤ Bk (B0 = 0).Additionally, if all but one of the input subblocksk-1 are wellk-1(b1,…,bk-1), then withprobability at least 1-Bk’ there is no logical effect and the output is wellk(b1,…,bk).
Dist-3 code setup
![Page 49: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/49.jpg)
Two operations:A.B. Error correctionC. (Logical) CNOT gate
Two indexed claims:A.B. Error correctionkC. CNOTk
Proofs by induction: Implications:
k
k
k
success except w/ prob.success except w/ prob.
Dist-3 code threshold proof
![Page 50: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/50.jpg)
CNOTkproof:
Assume input blocks are wellk(b1,…,bk). Declare failure if either Correctionkfails, or if there are two level k-1 failures.
On success, transverse CNOTsk-1 implement the correct logical effect (butpossibly correlate errors). The successful Correctionsk have no logical effect butrestore wellness (bounded dependencies).
Dist-3 code threshold proof Claim Bk (Correctionk): On success:
– Wellk(b1,…,bk) input ⇒ wellk(b1,…,bk) output,no logical effect– Arbitrary input ⇒ wellk(b1,…,bk) output
Failure prob. ≤ Bk (B0 = 0).Additionally, if all but one of input subblocksk-1 arewellk-1(b1,…,bk-1), then w/ prob. ≥ 1-Bk’, output iswellk(b1,…,bk) and no logical effect. Claim Ck (CNOTk): On success:
– Well inputs ⇒ well outputs, and logical CNOT– Arbitrary inputs ⇒ well outputs
Failure prob. ≤ Ck (C0 = p).
![Page 51: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/51.jpg)
CNOTk proof: Failure if either Correctionk fails, or if there are two level k-1 failures.
Success: transverse CNOTsk-1 implement correct logical effect. Correctionsk have nological effect.
Dist-3 code threshold proof Claim Ck (CNOTk): On success:
– Well inputs ⇒ well outputs, and logical CNOT– Arbitrary inputs ⇒ well outputs
Failure prob. ≤ Ck (C0 = p).
EC
EC
well
well
Dk
Dk
Dk
Dk
Dk
Dk
D1
D1
X
X
![Page 52: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/52.jpg)
Aliferis-Gottesman-Preskill threshold intuition
Aharonov & Ben-Or Idea: Maintain inductive invariant of (1-)goodness. (Ablock is good “if it has at most one bad subblock.”)
1. EC
ECX
X X
X
X
Two ways it can fail with distance-three codes:
2. EC
EC
X X
X
X
X
Both input blocks have a bad subblock. One input block has a bad subblock, andan additional error occurs.
![Page 53: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/53.jpg)
Aliferis-Gottesman-Preskill threshold intuition
EC
EC
well
well
well
well
A/B: Maintain ‘good’ness — two faults in rectangle cause logical failure (d≥5)
R: Maintain ‘well’ness — two faults in rectangle or well input cause logical failure
![Page 54: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/54.jpg)
Aliferis-Gottesman-Preskill threshold intuition
EC
EC
A/B: Maintain ‘good’ness — two faults in rectangle cause logical failure (d≥5)
R: Maintain ‘well’ness — two faults in rectangle or well input cause logical failure
A/G/P: two faults in extended (overlapping) rectangle cause logical failure
…errors in input come from errors inthe preceding error correction…
![Page 55: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/55.jpg)
Aliferis-Gottesman-Preskill threshold intuition
EC
EC
A/B: Maintain ‘good’ness — two faults inrectangle cause logical failure (d≥5) R: Maintain ‘well’ness — two faults inrectangle or well input cause logical failure
A/G/P: two faults in extended(overlapping) rectangle cause logicalfailure
…errors in input come from errors inthe preceding error correction…
![Page 56: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/56.jpg)
Aliferis-Gottesman-Preskill threshold intuition
EC
EC
A/B: Maintain ‘good’ness — two faults inrectangle cause logical failure (d≥5) R: Maintain ‘well’ness — two faults inrectangle or well input cause logical failure
A/G/P: two faults in extended(overlapping) rectangle cause logicalfailure
…errors in input come from errors inthe preceding error correction…
![Page 57: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/57.jpg)
Aliferis-Gottesman-Preskill threshold intuition
EC
EC
A/B: Maintain ‘good’ness — two faults inrectangle cause logical failure (d≥5) R: Maintain ‘well’ness — two faults inrectangle or well input cause logical failure
A/G/P: two faults in extended(overlapping) rectangle cause logicalfailure
…errors in input come from errors inthe preceding error correction…
![Page 58: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/58.jpg)
X Z
data
ancilla
mZ
mX
Steane-typeerror correction
mZ
mZap
ply c
orre
ctio
n
X
X
X
Teleportation
mZ
mX
U
data
ancilla
mX
mX
mZ
mZ
UL
Logi
cal
oper
atio
nsP
hysi
cal
oper
atio
nsKnill-type correction + computation
![Page 59: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/59.jpg)
Teleporting a CNOT gate
mZ
mX
mZ
mX
Logicaloperations
![Page 60: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/60.jpg)
Logicaloperations
Physicaloperations
Teleporting a CNOT gate
output blocks dependent!
![Page 61: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/61.jpg)
Logicaloperations
Physicaloperations
Teleporting a CNOT gate
assumeindependent
assumeindependent
independent!
⇒ Achieving independent errors on CNOT output blocks reduces to preparing encoded Bell states with block-independent errors
Unfortunately, this is impossible… But:
![Page 62: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/62.jpg)
Summary New threshold proof
– Based on bounding the distribution of errors in the system at eachtime step– More efficient than classical threshold proofs, leads to higher rigorousnoise threshold lower bounds– Works for concatenated distance-three codes
Possible extensions– Improved analysis of optimized standard fault-tolerance schemes
– Extend proof to work with schemes using distance-two codes andextensive postselection. Major difficulty is obtaining better control overerror distribution, particularly of dependencies and of errors in the badblocks.
(Ouyang, R.: 10-4)
![Page 63: Rigorous fault-tolerance thresholds · Stabilizer op. fault-tolerance Universal fault-tolerance Recent results Magic states distillation [Bravyi & Kitaev ‘04, Knill ‘04] – Universality](https://reader033.vdocuments.mx/reader033/viewer/2022042521/5f4cb34809b5fa18f7093be3/html5/thumbnails/63.jpg)
Blank slide