my experience is based on - mext.go.jp nsf support as a percentage of total federal support of...

30
US Federal Support for Mathematical Sciences – An NSF Perspective University of Tokyo 22 February 2010 Tony F Chan President, HK University of Science &Technology 2 My Experience is based on: Assistant Director for Directorate of Math & Physical Sciences, NSF 2006-2009 Dean of Physical Sciences, UCLA 2001-2006 Director, Institute for Pure & Applied Math, UCLA (NSF supported) 2000-2001 Chair, Math Dept, UCLA 1997-2000 NSF grantee for close to 30 years AMS, SIAM boards and committees None of this, including US system, may apply in Japan 3 Why Should Society Support Math Sci Research? Government: Investment in basic science, with long term return for society National needs: economy, health, energy, climate, defense, disasters Globally competitive S&T workforce Intellectual achievements alone insufficient “Interdisciplinary” necessary, but also insufficient Industry: Basic R&D, supports other units in company Mathematicians is a good talent pool, versatile. Private Philanthropy: Intellectual interest; “truth”, “beauty”, “applicability” Appreciates one of highest forms of human achievements Legacy, history Math is inexpensive: good return on investment!! 4 Math has an Image Problem Math is hard! Math is removed from everyday life. Math is static. Mathematicians --- super-smart but …. Math is subsumed in other popular terms: “software”, “algorithms ”, “computers”,… Mathematicians have no monopoly on the practice of math --- everyone (especially in S&T) does it. Math is most often “behind-the-scene” 5 US Government Support for Math Sci National Science Foundation (DMS, CISE,…) 46% Only agency supporting basic math sci research Primarily source of $ for academic research Dept of Defense (AFOSR, ARO, ONR, DARPA) 20% Dept of Energy (ASCR, SciDAC) 18% National Institute of Health (NIGMS, NIBIB,…) Intelligence agencies: NSA, NGA, CIA,.. Total = $537M in FY2010 6 Trends in Basic Research by Agency FY 1975-2007 OLPA-26 -389-

Upload: vocong

Post on 17-Mar-2018

220 views

Category:

Documents


6 download

TRANSCRIPT

Page 1: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

US Federal Support for Mathematical Sciences – An NSF Perspective

University of Tokyo22 February 2010

Tony F ChanPresident, HK University of Science &Technology

2

My Experience is based on:• Assistant Director for Directorate of Math & Physical

Sciences, NSF 2006-2009• Dean of Physical Sciences, UCLA 2001-2006• Director, Institute for Pure & Applied Math, UCLA

(NSF supported) 2000-2001• Chair, Math Dept, UCLA 1997-2000• NSF grantee for close to 30 years• AMS, SIAM boards and committees

• None of this, including US system, may apply in Japan

3

Why Should Society Support Math Sci Research?

• Government:– Investment in basic science, with long term return for society– National needs: economy, health, energy, climate, defense, disasters– Globally competitive S&T workforce– Intellectual achievements alone insufficient– “Interdisciplinary” necessary, but also insufficient

• Industry:– Basic R&D, supports other units in company– Mathematicians is a good talent pool, versatile.

• Private Philanthropy:– Intellectual interest; “truth”, “beauty”, “applicability”– Appreciates one of highest forms of human achievements– Legacy, history

• Math is inexpensive: good return on investment!!

4

Math has an Image Problem• Math is hard!• Math is removed from everyday life.• Math is static.• Mathematicians --- super-smart but ….• Math is subsumed in other popular terms: “software”,

“algorithms ”, “computers”,…• Mathematicians have no monopoly on the practice of

math --- everyone (especially in S&T) does it.• Math is most often “behind-the-scene”

5

US Government Support for Math Sci• National Science Foundation (DMS, CISE,…) 46%

– Only agency supporting basic math sci research– Primarily source of $ for academic research

• Dept of Defense (AFOSR, ARO, ONR, DARPA) 20%• Dept of Energy (ASCR, SciDAC) 18%• National Institute of Health (NIGMS, NIBIB,…)• Intelligence agencies: NSA, NGA, CIA,..

• Total = $537M in FY2010

6

Trends in Basic Research by Agency FY1975-2007

OLPA-26

-389-

Page 2: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

7

NSF Support as a Percentage of Total Federal Support of Academic Basic Research

(excluding NIH)

OLPA-29 8

The Political Budget Process• Presidential initiatives • Internal NSF budget process• OSTP, OMB• Congress Committees (hearings etc)• Lobbyists (hired by professional societies,

universities, etc)• Science community input (NRC etc)

9

The Fundamental Principle

• “If you expect the public and elected officials to support your research, you have an obligation to help them understand why it is in their – not your interest to do so” --- from Lewis Burke (a DC lobbying firm)

10

Agency Process for Developing Budget and Program Proposals

2009

11

Funding for Science & Engineering

DOD

Defense

Energy

Agriculture

Transportation

Education

Justice

Housing &Urban Development

EPA

NRC

NSF

Labor

NASA

Departments & Agencies

Engineering

PhysicalSciences

EnvironmentalSciences

LifeSciences

Psychology

SocialSciences

OtherSciences

Fields of Science

NationalDefense

AffairsInternational

Energy

Agriculture

Transportation

Health

Budget Review

Offices (OMB)

House & Senate Budget Committees (Budget Functions)

MilQual

Agriculture & Related Agencies

Science, Space, Justice, & Commerce

Energy and Water

Development

ForeignOperations

Interior

Labor, Health & Human Services, &

Education

Transportation & Related Agencies

Defense

House and Senate

Appropriations Subcommittees

Armed Services

Labor and Human Resources

Banking, Housing& Urban Affairs

Foreign Relations

Veterans Affairs

SenateAuthorization Committees

Energy & NaturalResources

Environment &Public Works

Commerce, Science,& Transportation

National Security

Economic & Educational

Opportunities

Banking & Financial Affairs

International Relations

Veterans Affairs

Commerce

Resources

Transportation & Infrastructure

Science

AgricultureAgriculture,

Nutrition, & Forestry

House Authorization Committees

(With significant R&D $)

Judiciary Judiciary

International Science,

Engineering and Technology

International Science,

Engineering and Technology

National Security National Security

Science Science

Technology Technology

Math & Computer Science

Agency for International Development

DOC/NOAA

Health & Human Services

Interior

National Security &

International Affairs

Natural Resources, Energy, and

Science

Economics &

Government

HumanResources, Veterans,and Labor

General Science, Space & Technology

Natural Resources & Environment

Commerce & Housing Credit

Community & Regional Development

Education, Training, Employment, & Social

Services

Veterans Benefits & Services

Administration of Justice

National Science and Technology

Council Research Committees

Environment and Natural

Resources

Environment and Natural

Resources

12

Community Inputs• Decadal surveys• NRC Studies• Interagency Committees/Working Groups (AAAC, HEPAP,

NSAC, BPA,…)• Professional Societies, University Groups• NSF Workshops, panels• MPS Advisory Committee• Committee of Visitors• Outreach: NSF staffs visit universities, confs

• Math Sciences does relatively little of this.

-390-

Page 3: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

13

Example of Recent Community Report and Presidential Initiative

•• Increase US talent poolIncrease US talent pool

•• Strengthen basic researchStrengthen basic research

•• Develop, recruit & retain best/brightestDevelop, recruit & retain best/brightest

•• Ensure innovation in AmericaEnsure innovation in America

•• From fundamental discoveries to marketable technologies.From fundamental discoveries to marketable technologies.

•• Facilities and instrumentationFacilities and instrumentation

•• World class science and engineering workforceWorld class science and engineering workforce

•• Focus on Phys Focus on Phys SciSci & Engineering& Engineering

•• Doubles NSF, DOEDoubles NSF, DOE--OS, NIST over 10 yearsOS, NIST over 10 years

14

OCI Example Cyber-enabled science: key part of NSF strategic plan

• Mathematical and Physical Sciences and Their Cyberinfrastructure Needs, 2004

• CHE Cyber Chemistry Workshop, 2004• Computation as a Tool for Discovery in Physics, 2002• Cyber infrastructure for the Atmospheric Sciences in the 21st Century,

2004• Petascale Computing in the Biological Sciences, 2006

Supporting Workshops/Reports:

15

Two Import Math Sci Reports

“If we wake up to discover that we have allowed the dominant position of U.S. mathematics to erode, we will pay a heavy price in foregone progress in technology, science, and economic productivity. Only if policymakers, legislators, and the mathematics community understand this danger alike can they act effectively to avoid it. At least two actions are clearly urgent: (a) providing more resources for mathematics and (b) using resources in more effective ways.”

William Odom, Lt. Gen. (Ret)Former Director of NSA

– Renewing U.S. Mathematics: Critical resources for the Future (Edward David, 1984)

– Report of the Senior Assessment Panel of the International Assessment of the U.S. Mathematical Sciences (William Odom, 1998)

16

Which model best for Math?

17 18

-391-

Page 4: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

19 20

21 22

Science the Endless Frontier : NSF

“The Government should accept new responsibilities for promoting the flow of new scientific knowledge and the development of scientific talent in our youth. These responsibilities are the proper concern of the Government, for they vitally affect our health, our jobs, and our national security.”

“Basic research is a long term process – it ceases to be basic if immediate results are expected on short term support.”

Vannevar Bush 1945

23

FY 2009 NSF R&RA Request

FY 2008 FY 2009Estimated Request Amount Percent

BIO $608.54 $612.02 $675.06 $63.04 10.3%CISE 526.68 534.53 638.76 104.23 19.5%ENG (less SBIR/STTR) 521.33 527.50 632.33 104.83 19.9% SBIR/STTR 108.67 109.37 127.00 17.63 16.1%GEO 745.85 752.66 848.67 96.01 12.8%MPS 1,150.73 1,167.31 1402.67 235.36 20.2%SBE 214.54 215.13 233.48 18.35 8.5%OCI 182.42 185.33 220.08 34.75 18.8%OISE 40.36 41.34 47.44 6.10 14.8%OPP 438.43 442.54 490.97 48.43 10.9%IA 219.45 232.27 276.00 43.73 18.8%US Arctic Research Comm. 1.45 1.47 1.53 0.06 4.1%Total, NSF $4,758.45 $4,821.47 $5,593.99 $772.52 16.0%

(Dollars in Millions)

Totals may not add due to rounding.

Change overFY 2008 EstimatedFY 2007

Actual

ACI

24

FY 2009 Budget Request by Division

FY 2008 FY 2009Es timated Req u es t A mo u n t Percen t

A s tro n o mical Scien ces $215.39 $217.86 $250.01 $32.15 14.8%Ch emis try 191.22 194.22 244.67 50.45 26.0%M ateria ls Res earch 257.27 260.22 324.59 64.37 24.7%M ath ematical Scien ces 205.74 211.79 245.70 33.91 16.0%Ph y s ics 248.47 250.52 297.70 47.18 18.8%M u ltid is c ip lin ary A ctiv ities 32.64 32.70 40.00 7.30 22.3%To tal, M PS $1,150.73 $1,167.31 $1,402.67 $235.36 20.2%

M athe matical and Phys ical Scie nce s Funding(Do llars in M illio n s )

T o t als m ay n o t add due t o ro un din g.

Ch an g e o v erFY 2008 Es timatedFY 2007

A ctu al

NSF:

$6.854 B,

+13.0%

-392-

Page 5: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

25

2011 NSF Budget Request: On the Doubling Path

26

27 28

29 30

MPS Ten-Year Funding HistoryMPS Subactivity Funding

(Dollars in Millions)

$0

$50

$100

$150

$200

$250

$300

FY98 FY99 FY00 FY01 FY02 FY03 FY04 FY05 FY06 FY07 FY08

DMR PHY CHE AST DMS OMA

-393-

Page 6: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

31 32

33

Research infrastructure

• Mathematical Sciences Institutes– ARCC, IAS, IMA, IPAM, MBI, MSRI, SAMSI– BIRS, IHES, MFO (Oberwolfach)

• CBMS conferences• SCREMS

– Scientific Computing Research Environments for the Mathematical Sciences

• MRI– Major Research Instrumentation

• et cetera

34

July 1 1979 NYT

New York Times 7/1/79

35

DMS’ Guiding principles

• Understand wider context of DMS• Articulate priorities to build budgets• Balance portfolio

36

Context: NSF investment prioritiesPromote transformational, multidisciplinary research

Investigate the human and social dimensions of knowledge and technology

Further U.S. economic competitiveness

Foster research that improves our ability to live sustainably on Earth

Advance fundamental research in computational science and engineering, and in fundamental, applied and interdisciplinary mathematics and statistics

-394-

Page 7: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

37

DMS investment priorities• Discovery

– advance the frontiers of research

• Connections– between sub-disciplines of mathematics and statistics– among mathematical and physical, biological, engineering, and

social sciences

• Community– cultivate a rich and diverse community of researchers, teachers,

students, knowledge workers, and citizens

38

Balance is delicateTraining: of postdocs, graduate students, undergraduates

Outreach: to other disciplines, decision makers, business leaders

Research: by faculty,postdocs, students

Breadth: intellectual scope, diversity of institutions, age, ethnicity, gender, geography

Education: of undergraduates,teachers, public

Depth: of ideas, strongest groups, strongest individuals

39

Balance is dynamicEmerging areas in the mathematical sciences

Opportunities at the interfaces with other disciplines

Core areas of themathematical sciences

Training the next generation of mathematical scientists

Supporting establishedresearchers engaged in cutting-edge work

Opportunities at the interfaces between core areas

40

Proactive Actions by Math Sci Community• Understand and play in the political process

– SIAM Sci Policy Committee; AMS DC Rep Sam Rankin• Articulate relevance of math sci to society• Present a unified and consistent message• Be “inclusive” in what is Math Sci• Partner with other disciplines• Emphasize intellectual depth, enabling nature to other

fields and workforce training• Consider international benchmarking• Cultivate public interest in math sci

41 42

-395-

Page 8: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

43

What’s Math got to do it?• Hedge funds ( )

• Google, GPS ( ), JPEG2000, MP3, cell phones…• E-commerce (e.g. RSA )• Titanic, Perfect Storm, Terminator 3, Cars,…• Medical imaging: CAT ( ) , MRI ( ),

PET( ),…• Recent Nobel Prizes: 1997 Economics (Finance), 2002

Chemistry (NMR ), 2003 Medicine (MRI), 2009 Physics (fibre optics)

• Landlocked Switzerland winning the America Cup

44

Math in the Media: Public is fascinated by Math

45

Wall Street Journal 1/29/09

46

The End – Thank you!

And Good Luck!

-396-

Page 9: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

- -

CREST

- Feb.23, 2010

(2009)

••••••

A BIG CHALLENGE

••

– , ..

• -––

•––

•– vs

• 4 1 180

••••

Jared Diamond

•1.

2.

3.

4.

� 1. � 4.

-397-

Page 10: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

JST Basic Research Programs

PRESTO(Precursory Research for Embryonic Science and Technology)

PRESTO(Precursory Research for Embryonic Science and Technology) aims to promote fundamental research in the field where a strategic focus is placed, as a part of a system that induces innovations leading to social and economic revolution, and to generate new innovative technology that leads to the development of scientific technology and the creation of new industries.

• 768-824

•–

–•

– Generalist Specialist

••• CREST

���

Sapporo Winter School , Feb.2009

-398-

Page 11: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

-399-

Page 12: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

HAHA3700 )

HA

(1968)

(2008)

,…

-400-

Page 13: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

CREST : JST's Funding Program for Team-based Basic Researches

2008 3 2009 5

2010!!

••••

•–

•–

•••

-401-

Page 14: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

© 2010 IBM Corporation

- -

IBM

© 2010 IBM Corporation

IBM Research - Tokyo

2 2010/02/23

� IT: 1,036

12363

( 2008 Annual Report)

IBM

1992 2000 2008

19%

3%

52%

19%

43%

6%

21%57%

17%11%

14%38%

(%)

: 1911

170

: 40

CEO: Samuel J. Palmisano

© 2010 IBM Corporation

IBM Research - Tokyo

3 2010/02/23

z/x POS

CRM ERP PLM VCM

Enterprise ApplicationEnterprise ApplicationEnterprise ApplicationBusiness Integration

z/OS

IBM

© 2010 IBM Corporation

IBM Research - Tokyo

4 2010/02/23

IBM Research

Zurich

Almaden

Tokyo

Haifa

19951995

19721972

198219821998199819551955

1995199519861986

India

19611961

ChinaWatson

Austin

© 2010 IBM Corporation

IBM Research - Tokyo

5 2010/02/23

IBM

WebSphere

© 2010 IBM Corporation

IBM Research - Tokyo

6 2010/02/23

(1) –

� 15� 80%

5 10SNS

RFID, IC

Source: TED 2007:Predicting the Next 5000 Days of the Web. IBM analysis

-402-

Page 15: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

© 2010 IBM Corporation

IBM Research - Tokyo

7 2010/02/23

(2)–

1 Chip

+

( )

( )

1985 1989

1.7*10^7

1.8*10^9

1997

2.0*10^11

2009

2.7*10^12

1988 1993 1998 2001

24,600

2241993 1997 2002 2006

(FRP at FAR=0.001)

0.79

0.01

0.540.20

312446

3,092

1954 // 1961 // 1973 //

1*10^51*10^6

3,092

1954 // 1961 5 // 1973 //

1*10^51*10^6

© 2010 IBM Corporation

IBM Research - Tokyo

8 2010/02/23

SoftwareProducts

MarketPositioningand Launch

DriveBusiness

20082007 20102009

Text Analytics, Optimization, Data Analytics, Modeling & Simulation, Security & Compliance, Stream Computing

Research

BAO Launch,Analytic Centers

Cognos, ILOG, SPSS, NISC

© 2010 IBM Corporation

IBM Research - Tokyo

9 2010/02/23

BAOBAO: Business Analytics and Optimization (2009 4 )

© 2010 IBM Corporation

IBM Research - Tokyo

10 2010/02/23

IBM

1970s 1980s 1990s 2000 2008

IBMSmarter Planet

© 2010 IBM Corporation

IBM Research - Tokyo

11 2010/02/2311

INSTRUMENTED

INTERCONNECTED

INTELLIGENT

Smarter Planet – IBM

10 1010000001© 20 IBM Cor

+

+=

Smarter Planet

© 2010 IBM Corporation

IBM Research - Tokyo

12 2010/02/23

Smarter Planet IBM 12 Smarter Planet

-403-

Page 16: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

© 2010 IBM Corporation

IBM Research - Tokyo

13 2010/02/23

Smarter Planet

Text Analytics- Natural language processing- Text Mining

Optimization- Mathematical Programming- Stochastic Optimization

Data Analytics- Anomaly Detection- Data Structure

Modeling & Simulation- Agent-based Simulation

Security & Compliance- Identity Resolution

Stream Computing- Stream Programming Language

etc.

© 2010 IBM Corporation

IBM Research - Tokyo

14 2010/02/23

Dynamic Programming

Succinct Data Structure

Mixed Integer Programming

Correlation Anomaly Detection

Hybrid Computingy p g

© 2010 IBM Corporation

IBM Research - Tokyo

15 2010/02/23 © 2010 IBM Corpor10/02/23

-“Collaboratory” = Collaboration + Laboratory

(Collaboratory)

© 2010 IBM Corporation

IBM Research - Tokyo

16 2010/02/23

IBM

IBM

-404-

Page 17: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

BTCC JST

1.2.3.4.

cf: Vol. 1, No. 1, 1991

19C

a1, , an

11 0n n

nx a x a�� � � � 0na� �a

1 2

1 2 3 1 2 3

1 2 3 4 1 2 32

3 1 2 3 1 4

2 0, 03 0, 0, 0, 04 0, 0, 0, 0, 0

( ) 0

n a an a a a a a an a a a a a a a

a a a a a a

� � � �� � � � � � �� � � � � � � �

� � �

n=3 C. Maxwell 1867n=4, 5 Maxwell

Routh (1872) Hurwitz (1877)

Hermite

Sturm

Lyapunov 1893

APA+ATP= I P

[ ]

( , ) : :( ) :

x f x u x uy g x y��

x f ( , )

uyx

( )u k x�

( )u k y�

: , ::

x Ax Bu A n n B n ry Cx C m n� � � �� �

x Ax BAx B

:u Ky K r n� �

� x A BKC x� ��x A BK�AA�

( )A A BKC K �

Eigenvalue Assignment

n λ1, λ2, λn

A+BKC K

C=I

1nrank B AB A B n�� � � �1nA B1nA B1n

-405-

Page 18: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

n m + r -1− 1

Kimura (1975)citation number 258

mr nmr m + r -1

R. Hermann

generic

n ≥ mr

Byrnes C.I.

Grassman

Schubert Calculus

1!2! ( 1)!1!2! ( 1)!( )!1!2! ( 1)!r m mr

m r� �

� �( 1)!1!2! ( 1)!( )1)!1!2! ( 1)!(1)!1!2! ( 1)!(1)!1!2! (1)!1!2! (

(

Wang, 1995

n ≥ mr-1

[ ]

( )K s

( ) :G s

( )G s

G(s) Δ(s)

G(s) + Δ(s)

Δ(s)K(s)

max ( ) ( ).j r j� �� � �

-406-

Page 19: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

N-P 1914-18

l α1, α2, , αl , ; β1, β2, , βl

Re[s]>0 |w|<1

f(s) αi, βi

� �Re 0, 1, 1,2, , .i i i l� �� � � , .l,

( ) , 1,2, ,i if i l� �� � l,

Schwarz

Pick Nevanlinna

, 1, ,

10i j

i j i j l

� �� �

� ���� �� ��� � l, ,

etc.

αi G(s)

βi

αi

( )( )

ii

i

rG���

�( )G

( )G s( )G( i��

N-P

• H∞

IEEE CSS George Axelby Award

Citation Number 146

etc.

etc.

-407-

Page 20: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

2010.2.23

( )

( )

( )

2010(Department of Economics)(i) (ii)

(iii) ( )

(i) Mas-Colell, Whinston and Green(1995) ”Microeconomic Theory”(Oxford), (ii) Ljungqvist and Sargent (2004) ”Recursive Macroeconomic Theory” (MIT UP), (iii) Hayashi (2000) “Econometrics” (Princeton UP)

(iv) Duffie (2001) “Dynamic Asset Pricing Theory” (Princeton UP)

von Neumann, S. Kakutani, K. Ito

1969

vs. vs.

( )

(Department of Statistics) 20101945

OBK. Ito

(derivatives)(SDE)

K. Ito

2010

BIGLOBE

-408-

Page 21: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

-409-

Page 22: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

-

2

� US PhD� Constrained Optimization, Dynamic Optimization� Separating Hyperplane Th)�

� PhD

� 1960 70

Outline

� 4�

3

12

55

22

-410-

Page 23: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

TLO

TLO

8

9

WG

���

����

-411-

Page 24: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

•••

•••

WG

2 1 1 3 5

H21.12.8

22

22

� 2020

6

H21.12.30

-412-

Page 25: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

話題提供

–パネルディスカッションのために –三井 斌友

(同志社大学理工学部数理システム学科)

MathSurvey 10 – p.1/4

ある事例 –関数概念をめぐって

機械工学専攻の院生の質問

MathSurvey 10 – p.2/4

ある事例 –関数概念をめぐって

機械工学専攻の院生の質問

熱工学研究室でのテーマ

MathSurvey 10 – p.2/4

ある事例 –関数概念をめぐって

機械工学専攻の院生の質問

1次元逆問題

MathSurvey 10 – p.2/4

ある事例 –関数概念をめぐって

機械工学専攻の院生の質問

1次元逆問題

0 < x < 1上で初期関数(温度分布)f0(x)から出発して

MathSurvey 10 – p.2/4

ある事例 –関数概念をめぐって

機械工学専攻の院生の質問

1次元逆問題

0 < x < 1上で初期関数(温度分布)f0(x)から出発して

fn+1(x) =

∫ x

0

K(ξ, x)fn(ξ) dξ (n = 0, 1, 2, . . .)

MathSurvey 10 – p.2/4

-413-

Page 26: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

ある事例 –関数概念をめぐって

機械工学専攻の院生の質問

1次元逆問題

0 < x < 1上で初期関数(温度分布)f0(x)から出発して

fn+1(x) =

∫ x

0

K(ξ, x)fn(ξ) dξ (n = 0, 1, 2, . . .)

によって,{fn(x)} の振舞いを観察したい.

MathSurvey 10 – p.2/4

ある事例 –関数概念をめぐって

機械工学専攻の院生の質問

1次元逆問題

0 < x < 1上で初期関数(温度分布)f0(x)から出発して

fn+1(x) =

∫ x

0

K(ξ, x)fn(ξ) dξ (n = 0, 1, 2, . . .)

によって,{fn(x)} の振舞いを観察したい.積分核 K(ξ, x) は指数関数的減少

MathSurvey 10 – p.2/4

ある事例 –関数概念をめぐって

機械工学専攻の院生の質問

1次元逆問題

0 < x < 1上で初期関数(温度分布)f0(x)から出発して

fn+1(x) =

∫ x

0

K(ξ, x)fn(ξ) dξ (n = 0, 1, 2, . . .)

によって,{fn(x)} の振舞いを観察したい.積分核 K(ξ, x) は指数関数的減少

二つの方向:

MathSurvey 10 – p.2/4

ある事例 –関数概念をめぐって

機械工学専攻の院生の質問

1次元逆問題

0 < x < 1上で初期関数(温度分布)f0(x)から出発して

fn+1(x) =

∫ x

0

K(ξ, x)fn(ξ) dξ (n = 0, 1, 2, . . .)

によって,{fn(x)} の振舞いを観察したい.積分核 K(ξ, x) は指数関数的減少

二つの方向:

(1)解析的な積分(computer algebra など)による

MathSurvey 10 – p.2/4

ある事例 –関数概念をめぐって

機械工学専攻の院生の質問

1次元逆問題

0 < x < 1上で初期関数(温度分布)f0(x)から出発して

fn+1(x) =

∫ x

0

K(ξ, x)fn(ξ) dξ (n = 0, 1, 2, . . .)

によって,{fn(x)} の振舞いを観察したい.積分核 K(ξ, x) は指数関数的減少

二つの方向:

(1)解析的な積分(computer algebra など)による

(2)数値的(近似的)積分による

MathSurvey 10 – p.2/4

ある事例 –関数概念をめぐって (2)

質問の要点:数値積分をどう実行するか

MathSurvey 10 – p.3/4

-414-

Page 27: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

ある事例 –関数概念をめぐって (2)

質問の要点:数値積分をどう実行するか

「関数の概念」が十分理解できていない

MathSurvey 10 – p.3/4

ある事例 –関数概念をめぐって (2)

質問の要点:数値積分をどう実行するか

「関数の概念」が十分理解できていない

x ∈ [0, 1] → fn(x)という写像としてとらえる

MathSurvey 10 – p.3/4

ある事例 –関数概念をめぐって (2)

質問の要点:数値積分をどう実行するか

「関数の概念」が十分理解できていない

x ∈ [0, 1] → fn(x)という写像としてとらえる

これは同時にプログラミング(ソフトウェア)にも生きる概念

MathSurvey 10 – p.3/4

ある事例 –関数概念をめぐって (2)

質問の要点:数値積分をどう実行するか

「関数の概念」が十分理解できていない

x ∈ [0, 1] → fn(x)という写像としてとらえる

これは同時にプログラミング(ソフトウェア)にも生きる概念

xを入力すれば f(x)が出力される

MathSurvey 10 – p.3/4

ある事例 –関数概念をめぐって (2)

質問の要点:数値積分をどう実行するか

「関数の概念」が十分理解できていない

x ∈ [0, 1] → fn(x)という写像としてとらえる

これは同時にプログラミング(ソフトウェア)にも生きる概念

xを入力すれば f(x)が出力される

[0, 1]を等間隔分割 xk = kΔx (k = 0, 1, 2, . . .)して

MathSurvey 10 – p.3/4

ある事例 –関数概念をめぐって (2)

質問の要点:数値積分をどう実行するか

「関数の概念」が十分理解できていない

x ∈ [0, 1] → fn(x)という写像としてとらえる

これは同時にプログラミング(ソフトウェア)にも生きる概念

xを入力すれば f(x)が出力される

[0, 1]を等間隔分割 xk = kΔx (k = 0, 1, 2, . . .)して

fn+1(xk) =k−1∑j=0

K(ξj , xk)fn(ξj)Δx

MathSurvey 10 – p.3/4

-415-

Page 28: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

ある事例 –関数概念をめぐって (2)

質問の要点:数値積分をどう実行するか

「関数の概念」が十分理解できていない

x ∈ [0, 1] → fn(x)という写像としてとらえる

これは同時にプログラミング(ソフトウェア)にも生きる概念

xを入力すれば f(x)が出力される

[0, 1]を等間隔分割 xk = kΔx (k = 0, 1, 2, . . .)して

fn+1(xk) =k−1∑j=0

K(ξj , xk)fn(ξj)Δx

数列 {fn(xk)}から {fn+1(xk)}からへの写像MathSurvey 10 – p.3/4

ある事例 –関数概念をめぐって (3)

こんな簡単な計算で済むか?

MathSurvey 10 – p.4/4

ある事例 –関数概念をめぐって (3)

こんな簡単な計算で済むか?

もちろん No!

MathSurvey 10 – p.4/4

ある事例 –関数概念をめぐって (3)

こんな簡単な計算で済むか?

もちろん No!

これから出発して,より高度なシミュレーションに移ることがで

きるだろう

MathSurvey 10 – p.4/4

-416-

Page 29: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

����

�����

�������

����������������� !"�#$%�&'()*+,-�./01�%2345�1678,9:�;<:=

(>��?@A;BC *D1�EF?

�'!"

• �GH�I�JK'L,*M0�NO4PQ!"*+�RS5T;UVW.X1=

• YZ[\']^?\_,`a���bc4d:efg'�.�Uh ij�1=

• kl�mn�op�qrstu�vwc'xy�.�RS5� z{0.�:=

• �E�(>!";�c#$�'|,��}~��|�b��U,V15�g�b�8���,��.�E5�0�5�5�g'�{0.�:=

��_������ MRIqrMRI��

(>'�

• (>;�����X����� ���0����5@;� ¡�4¢�.�1=

• £H�¤¥¡<W1=• K¦§¨ (>�l©ª4�«¬'�1'X�A�A' UV­�®¯;°±�²³,´�µ|A' ¶·1=

• ¸¹º»¯,��¼½;¾* ¿W11Q�¼ÀÁ �4ÂÃ:ÄÅ'5a��Æ�4ÇÈ�1=

��Æ�É �<:ʹ¶·:'�¬¾ ËÌ�1=A0 Í0E�.�:=

1867Î ÏÐÑ�ÒÓJ1872Î ÔÕÖ1986Î ×��ØÕÖ1989Î �(>×�ÙÚ�Ö1895Î (�ÛÜ,ݵ1905Î (ÞÛÜ,ݵ

ßÎà�áâãäÎ�åæçè��éêäëì�íêáãì ¡<W1=

�,µ�.�BC

• b����� �î�ï�U,V15�g'�ðñ (>;�òó%;<:=

• ô�õ467�.�1ö��b�÷ø�A'4ù�:���úû;<:g'��01=

• 1970Îü,¸¹NOº»,ý�þ�Ç �{01 �����Ȫ,��þ�Ç��5|����,��aó�05EW1=

• ��¾����ü`a�HÅ���'xy�1�����4�C�.�1=

��4��1 �`|E�5EW1=�¾�;�������4Q��1F����ö�����W.�1=

� A�R5ù�4 Q1E��5�=

¸!"¦(1875.�4.�21�#1960.�2.�28)

• $%&_$_�,'��([)[\�*;ô�õ�674+P�1=

• 1901Î�,�à��-×��;�1=

• .1/î0�Û;1?�%2�34 5E5|5W1�46Å,�÷7E�ù�.�167,89��:;õ4<=�1=

• �>?,¡@�;�CABõÈlî�CD4PQE��>4��1=

• ¸!��(>'�F5a�1?�%2;���� G�5A'4�W.�1=

¸!"¦�HaIX��E JKW1A'4 Q1F

-417-

Page 30: My Experience is based on - mext.go.jp NSF Support as a Percentage of Total Federal Support of Academic Basic Research (excluding NIH) OLPA-29 8 The Political Budget Process † Presidential

委託業務成果報告書の無断複製等禁止の標記について

委託業務に係る成果報告書の無断複製等の禁止の標記については、次によるものとする。

本報告書は、文部科学省の平成 21 年度科学技術調査資料作成委託事業による委託業務とし

て、国立大学法人九州大学、国立大学法人東京大学、社団法人日本数学会、新日本製鐵株

式会社が実施した「数学・数理科学と他分野の連携・協力の推進に関する調査・検討~第

4期科学技術基本計画の検討に向けて」の成果を取りまとめたものです。

従って、本報告書の著作権は、上記 4 機関に帰属しており、本報告書の全部又は一部の無

断複製等の行為は、法律で認められたときを除き、著作権の侵害にあたるので、これらの

利用行為を行うときは、上記 4機関の承認手続きが必要です。

但し、文部科学省が求めた場合は委託調査の結果を無償で使用することを許可します。ま

た、引用など使用する際には、委託業務の成果であることを明示することとします。