Computational Thinking

and
CS@CMU

Jeannette M. Wing
President’s Professor and Head
Computer Science Department
Carnegie Mellon University

Microsoft Faculty Summit, Hangzhou, China, October 31, 2005

 My Grand Vision for the Field
• Computational thinking will be a fundamental skill used
by everyone in the world by the middle of the 21st
Century.

– Just like reading, writing, and arithmetic.

– Imagine every child knowing how to think like a computer
scientist!

– Incestuous: Computing and computers will enable the spread of

computational thinking.

CSD 2 Jeannette M. Wing

 Examples of Computational Thinking
• How difficult is this problem and how best can I solve it?
– Theoretical computer science gives precise meaning to these and related questions and
their answers.
• C.T. is thinking recursively.
• C.T. is reformulating a seemingly difficult problem into one which we know how to
solve.
– Reduction, embedding, transformation, simulation
• C.T. is choosing an appropriate representation or modeling the relevant aspects of
a problem to make it tractable.
• C.T. is using abstraction and decomposition in tackling a large complex task.
• C.T. is judging a system’s design for its simplicity and elegance.
• C.T. is type checking, as a generalization of dimensional analysis.
• C.T. is prevention, detection, and recovery from worst-case scenarios through
redundancy, damage containment, and error correction.
• C.T. is modularizing something in anticipation of multiple users and prefetching
and caching in anticipation of future use.
• C.T. is calling gridlock deadlock and avoiding race conditions when synchronizing
meetings.
• C.T. is using the difficulty of solving hard AI problems to foil computing agents.
• C.T. is taking an approach to solving problems, designing systems, and
understanding human behavior that draws on concepts fundamental to computer
science.
Please tell me your favorite examples of computational thinking!
CSD 3 Jeannette M. Wing
 Evidence of Computational Thinking’s Influence
• Computational thinking, in particular, machine learning has
revolutionized Statistics
– Statistics departments in the US are hiring computer scientists
– Schools of computer science in the US are starting or embracing
existing Statistics departments
• Computational thinking is CMU’s current big bet in Biology
– Algorithms and data structures, computational abstractions and
methods will inform biology.
• Computational thinking is CMU’s next big bet in
– Game Theory
• Electronic marketplaces, multi-agent systems, security, and networking
– Nanocomputing
• Molecular-scale computing based on reconfigurable fabric makes the
chemistry easier.

CSD 4 Jeannette M. Wing

 Analogy
The boldness of my vision: Computational thinking is
not just for other scientists, it’s for everyone.

• Ubiquitous computing was yesterday’s dream, today’s

reality
• Computational thinking is today’s dream, tomorrow’s
reality

CSD 5 Jeannette M. Wing

 Computational Thinking
• Conceptualizing, not programming
– Computer science is not just computer programming
• Fundamental, not rote skill
– A skill every human being needs to know to function in
modern society
– Rote: mechanical. Need to solve the AI Grand Challenge of
making computers “think” like humans. Save that for the
second half of this century!
• Ideas, not artifacts
– It’s not just the software and hardware that touch our daily
lives, it will be the computational concepts we use to
approach living.
• It’s for everyone
– C.T. will be a reality when it is so integral to human
endeavors that it disappears as an explicit philosophy.
CSD 6 Jeannette M. Wing
 Computer Science at Carnegie Mellon
CS@CMU

CSD 7 Jeannette M. Wing

 Computing at Carnegie Mellon
CMU

Fine Arts Social Sciences Science School of Engineering Business Public Software
Design Psychology Biology Computer Mechanical Policy Engineering
Drama Philosophy Math Science Electrical Institute
Statistics
BS
PhD MS PhD MS PhD PhD MS
Human Center for Computer Robotics MD/PhD
Computer Automated Science MS Institute (RI)
Interaction Learning and Department
Institute (HCII) Discovery (CALD) (CSD) 2 PhD 4 MS
PhD 2 MS Institute for Software
MS Engineering, International
Entertainment Language
Technology Technologies (ISRI) Distance
Center (ETC) Institute (LTI) PhD
Neural Cognition Supercomputing
Medical
Linguistics
3 MS BS
West Coast Campus Qatar Campus
Pitt

CSD 8 Jeannette M. Wing

 SCS Numbers at a Glance
• 215 faculty
• 213 courses on the books
• 540 bachelors students
– including a handful of HCI double majors
• 235 masters students across 11 programs
• 400 doctoral students across 9 programs

CSD 9 Jeannette M. Wing

 CS@CMU Distinguishing Characteristics

• Research Style
• Leadership in Education
• Supportive Culture
• Women in Computer Science
• Organizational Structure

CSD 10 Jeannette M. Wing

 Research Style

• Broad vision of computer science

• High quality, high impact
• Collaborative, interdisciplinary
– Cross styles: Systems, Theory, AI
• We build things. For real users. Systems ⇔ Theory
• We think big.
• We take risks.

CSD 11 Jeannette M. Wing

 CSD Collaboration Network [Carley 2004]

CSD 12 Jeannette M. Wing

 Leadership in Education

• Ph.D. Program
– Research from Day One
– Immigration Course, Handshake
– Speakers Club
– Black Friday
• Undergrad program
– Challenging and unique curriculum
– Devoted faculty: University advising and teaching awards
– Senior faculty teach large core courses and upper-level electives
• Elite 5th Year MS
– For our own best and brightest
– To encourage early exposure to research

CSD 13 Jeannette M. Wing

 Supportive Culture

• Reasonable Person Principle

– Assume everyone around you is reasonable, including yourself
– Implies a high level of mutual trust, support, and respect
• Collective responsibility
– Black Friday: all faculty take responsibility for all graduate
students
– Annual SCS Teaching Meeting: Open sanity check on teaching
assignments
• Presume success
– If we admit you, we believe you are capable of finishing.
– If we hire you, we expect you to succeed up the promotion
ladder.

CSD 14 Jeannette M. Wing

 Women in Computer Science

• 39% of our 2000 enrollment was female

• 33% of our graduates on average are female
– Twice the national average in BS degrees

• Women@SCS has changed the culture at CMU

– Broken the gender divide
– Actively engaged in community within and outside CMU
• Mentoring, tutoring, outreach
• Computer Professionals for Social Responsibility
• SCS Day

CSD 15 Jeannette M. Wing

 Unusual Organizational Structure
• Our model of CSD and SCS is of an expanding
universe.
– CSD is
• Home for traditional areas of Computer Science
• Home for new, emerging areas of Computer Science
– Other units of SCS connect Computer Science to other
disciplines
• Robotics: CS + Mechanical Engineering + Electrical Engineering
• Language Technologies: CS + Linguistics
• Human-Computer Interaction: CS + Design + Psychology
• Automated Learning and Discovery: CS + Statistics
• Software: CS + Public Policy + Management
• Lack of administrative boundaries
– Unit heads get along and we get along with the Dean
– Not a department head-dean adversarial relationship.

CSD 16 Jeannette M. Wing

 What Do We Do?

CSD 17 Jeannette M. Wing

SCS’s Research Enterprise
New, emerging areas (e.g., CS + X):
• Computational biology
• Computational astrophysics
• Nanocomputing
• Foundations of privacy
• Computing technology and society
RI •… LTI
AI: robotics, vision AI: natural language processing, speech
CSD
New, emerging
areas in Theory: Theory: algorithms, complexity, semantics New, emerging areas in
• game theory Systems: computer architecture, O/S, Systems:
•… distributed systems, networking, databases,• pervasive computing
performance modeling, graphics, • trustworthy computing
New, emerging areas
programming languages, formal methods • post-Moore’s Law
in AI:
AI: planning, learning, search, cognition, computers
• optimization
computational neuroscience •…
• coaching
•…
Systems: software engineering, Systems: human-computer interfaces
public policy, e-commerce
AI: machine learning HCII
ISRI
CSD 18
CALD Jeannette M. Wing
 What We Do: Research in CSD
• Algorithms and Complexity
• Artificial Intelligence
• Computational Molecular
Biology
• Computational Neuroscience
• Computer Architecture
• Databases
• Formal Methods
• Graphics
• Human-Computer
Interaction
• Large-Scale Distributed
Systems
CSD 19
• Machine Learning
• Mobile and Pervasive
Computing
• Networking
• Principles of Programming
• Robotics
• Scientific Computing
• Security
• Software Engineering
• Technology and Society
• Vision, Speech, and Natural
Languages
Jeannette M. Wing
 Three Mosaics

CSD 20 Jeannette M. Wing

Some Highlights Applied Math Models
Game theory
Quantifiable Results
Automated mechanism design applicable to divorce
Using deconvolution to correctly identify settlements and tie-breaking rules [Sandholm]
15% more cycling genes in yeast cells Near optimal on-line auctions [A. Blum]
when compared to using observed
values alone [Bar-Joseph] Predict Internet stability wrt congestion control
if end-points act selfishly [Seshan]
A new search algorithm to solve the k-nearest
neighbor problem with a 10-fold speedup over the Knot theory
best metric-tree algorithm [Moore]
A new topological approach to
A new data structure for representing n-vertex detecting protein similarity leading
unlabeled graphs using O(n) bits and supporting to a representation of proteins by
adjacency and degree queries in constant time line weavings [Erdmann]
[Blelloch]

Metric spaces
A new spike representation of auditory
signals resulting in a coding 3x more Complexity of metric spaces and
efficient than MP3 [Lewicki] applications to TSP-like problems,
networking, web-page clustering [Gupta]

Understanding Intelligence
Micro Macro
Theory of consciousness Multi-agent (robots and
[M. Blum, Rudich, A. Blum] humans) planning and Multi-participant (robots and
learning [Veloso] humans) dialog and
Use fMRI data to conversation [Rudnicky]
construct better cognitive
models [Mitchell]

Neural representation of space

(Where am I?) in rats (and robots)
[Touretzky]
CAPTCHA, ESP:
Role of feedback from higher visual areas Using hard AI
on early (V1 and V2) areas by studying problems to solve
CSD awake behaving monkeys [T.S. Lee] 21 crypto [M. Blum] Jeannette M. Wing
Some Common Themes
Lots of data
Texture synthesis uses Shannon’s N-grams
info theoretic technique to quilt radishes,
rocks, and yogurt. [Efros]
Fractals and power laws to model sensor
data, network graphs, multimedia data,
protein interactions. [Faloutsos]
Use server logs from content-delivery
networks to estimate interdomain Web
traffic flow. [Maggs]
Manage distributed
Anomaly detection based on real
dynamic data applied to
network data [Maxion, Tan]
Web monitoring [Olston]
Multispectral imagery
+ photogrammetric knowledge
+ large-scale databases = digital maps
[Cochran, McKeown]
Machine learning
For many, many things [Bar-Joseph, A. Blum, Efros,
Lafferty, Langmead, Lewicki, Mitchell, Moore,
Sandholm, Veloso]
Probability and
statistics SYNC: Scheduling Your Network
Connections [Harchol-Balter]
Distributed inference in sensor networks [Guestrin]
Generalized Chernoff bounds on
event probabilities for graphical
CSD
models [Lafferty]
Data-Driven Sparse data
With sparse experimental
data (to minimize wet lab
cost), determine 3-D
structures and dynamics
of nucleic acids and
proteins [Langmead]
Making use of labeled and
Use motion capture data to render human unlabeled data [A. Blum, Lafferty]
behavior efficiently [Hodgins]
Use precomputed data-driven deformable object simulation
for computer animation, video games, reality based
modeling, manufacturing and tissue simulation. [James]
Type theory
ConCert: grid computing
[Crary, Harper, Lee, Pfenning]
Secret Weapons Separation logic
For concurrency [Brookes, Reynolds]
Model checking
Software architecture
Rainbow: runtime adaptation of self-
For hybrid systems, software, and managing systems [Garlan, Schmerl,
security [Bryant, Clarke, Wing] Steenkiste]
Economics, decision theory
Value-driven software engineering [Shaw]
E-commerce, voting, auctions [Sandholm]
22 Jeannette M. Wing
Some Big and Wild Projects Large/Integrative Systems
Macro
Interdisciplinary/Collaborative 100 Mbps to 100 million homes [Zhang]

(Surprises) Outside CSD Internet-Suspend-Resume on

campus [O’Hallaron, Satya]
Simulating blood flow, with
computational fluid dynamicists, PCtvt [Reddy]
hemorheologists, …, transplant Informedia->Caremedia->
surgeons [Miller, Blelloch] Quality of Life Technology
Institute: care of elderly and
chronically disabled and ill
Use of convolution integrals for modeling Humanoid robots [Christel, Gao, Hauptmann,
super-secondary structures in proteins, [Hodgins, Kanade] Ng, Wactlar]
with Pitt biologist [Carbonell]

Micro
Safety of adaptive cruise control,
with ECE; of insulin pump, with
Chem Eng [Clarke] Claytronics: Synthetic reality
RADAR/CALO [Carbonell, through programmable matter
Fahlman, Fink, Moore, [Goldstein, Mowry]
Rudnicky, Siewiorek, Veloso]
Astronomy databases, with
astrophysicists [Ailamaki]
Tablet PCs in
Slashdot/Wired: Past and Future?
15-100, with Simulating trees blowing in the wind [James]
HP and MS
[Guna] Two robot hands shaking
hands [Pollard]

Analysis of multiserver
systems, with Tepper
[Harchol-Balter] Robot team
building a 4-beam Unmanned aerial vehicles Query-by-Humming
structure in space [Kanade, Ke, Veloso] [Dannenberg]
Within CSD
[Simmons]
Theorem-proving cell phones
Origami robots [Mason]
[Pfenning, Reiter] Building virtual worlds [Pausch]

Model checking Proof-Carrying Code

[Clarke, P. Lee] 3-D Quake [Seshan]
CSD 23 Jeannette M. Wing
 Educational Portfolio in SCS at CMU
• 7+5 Ph.D. programs
– Major: Computer Science, Robotics, Language Technologies,
Human-Computer Interaction, Software Engineering,
Computational Statistics, Computers, Organizations, and Society
– Joint/Special: Algorithms, Complexity, and Optimization,
Computational Biology, Neural Basis of Cognition, Pure and Applied
Logic, Ph.D./M.D. with Univ. of Pittsburgh
• 11 M.S. programs
– Professional, e.g., Software Engineering, Information Technology,
Human-Computer Interaction, Entertainment Technology
– Academic, e.g., Robotics, Language Technologies, Fifth-Year
Master’s in Computer Science
• 1 B.S. program
– Taught primarily by CSD faculty

CSD 24 Jeannette M. Wing

 What Computer Science Is and Is Not
IS
• Alan Newell’s view
– “Computer science is the study of all the phenomena arising from
[computers].”

IS NOT
• Public misperception
– Computer science is computer programming. NOT!
– Computer science is computational thinking.
• Government funding agencies misperception
– Computer science is over. The rest is engineering. NOT!
– Computer science is driven by technology trends.

IS
• Computer science is limited only by our imagination!

CSD 25 Jeannette M. Wing

 Grand Vision for the Field
• Computational thinking will be a fundamental skill used
by everyone in the world by the middle of the 21st
Century.

• Join us at Carnegie Mellon toward making computational

thinking commonplace!

CSD 26 Jeannette M. Wing