Transcript ppt - Computer Science, Columbia University

Introduction to Artificial
Intelligence
Prof. Kathleen McKeown
722 CEPSR, 939-7118
TAs:
Kapil Thadani
724 CEPSR, 939-7120
Phong Pham
TA Room
Today



What is artificial intelligence
anyway?
Requirements and assignments for
class
Examples of AI systems
2
What is intelligence?

Intelligence
“The ability to learn and solve problems”
(Webster’s Dictionary)
 The ability to think and act rationally


Goal in artificial intelligence

Build and understand intelligent
systems/agents
3
2001
4
Definitions
Systems that think like Systems that think
humans
rationally
The exciting new effort to make
computers think .. Machines with
minds, in the full and literal sense
(Haugeland, 1985)
..systems that exhibit the
characteristics we associate with
intelligence in human behavior –
understanding language,
learning, reasoning, solving
problems and so on (Handbook of
AI)
Systems that act like
humans
Systems that act
rationally
The study of how to make
computers do things which, at
the moment, humans do better
(Rich and Knight)
..the study of [rational] agents
that exist in an environment and
perceive and act. (Russell and
Norvig)
6

Systems that think like humans
versus

Systems that act like humans
7

Systems that think rationally
versus

Systems that act rationally
8
Different Approaches to AI

Building exact models of human cognition


The logical thought approach


The view from psychology and cognitive
science
Emphasis on correct inference
Building rational agents


Agent: something that perceives and acts
Emphasis on developing systems to match or
exceed human performance, often in limited
domains
9
Class focus

Systems that act
Like humans
 Rationally

10
AI is a smorgasbord of topics

Core areas




Perception




Knowledge
representation
Reasoning/inferenc
e
Machine learning
Vision
Natural language
Robotics
Uncertainty


General algorithms




Applications






Game playing
AI and education
Distributed agents
Decision theory

Probabilistic
approaches
Search
Planning
Constraint
satisfaction
Electronic
commerce
Auctions
Reasoning with
symbolic data
11
AI is a smorgasbord of topics

Core areas




Perception




Knowledge
representation
Reasoning/inferenc
e
Machine learning
Vision
Natural language
Robotics
Uncertainty


General algorithms




Applications






Game playing
AI and education
Distributed agents
Decision theory

Probabilistic
approaches
Search
Planning
Constraint
satisfaction
Electronic
commerce
Auctions
Reasoning with
symbolic data
12
AI used to be

Expert systems
Medical expert systems – diagnosis
 Computer systems design



Theorem proving/software
verification
Inheritance, class-based systems
13
AI is interdisciplinary







Psychology
Cognitive Science
Linguistics
Neuroscience
Economics
Philosophy
Physics
14
What will we study in the course?
15
Assignments

2 programming assignments
Search (1.5 weeks)
 Game playing (3.5 weeks)
 Tournament


1 light programming/using tool plus paper
(3 weeks) – machine learning

1 purely written assignment (1 week)

Each programming assignment has
written questions too
16
Grading


45% homeworks – homeworks are
important. You can’t pass without doing
them.
5% class participation






Notes will be posted on the web
There will be board work in addition to slides.
The slides don’t tell the whole story.
Class is a social experience – there will be
discussion
End of Class Questions
20% midterm
30% final
17
Undergrad vs. MS

Separate grading curves

Separate game tournaments

MS students picked to raise
discussion issues; undergrads
expected to respond
18
Reading


Chapters from the required text:
Artificial Intelligence: A Modern
Approach, Russell and Norvig, 2003.
Columbia University Bookstore.
Selected papers. Watch for papers
on reserve.

Will be posted on the Reading Section of
the web
19
Other AI Classes this semester





4701 NLP (Hirschberg)
4731 Computer Vision (Nayar)
4737 Biometrics (Belhumeur)
6733 3D Photography (Allen)
6998 Section 4 Search Engine
Technology (Radev)
20
Some Examples

Natural language processing
Question answering on the web
 Automatic news summarization


Robotics
Robocup soccer
 Roomba: robotics meets the real world


Vision

Modeling the real world
21
Machine Learning

Learning to play pool

Talking robots
22
Today’s Assignment



Fill out on courseworks
Survey worth 5 points towards total homework
grade
Answer the following questions







UNI:
Degree: BA
BS
MS
PhD
non-degree
Year at Columbia (e.g., freshman, sophomore,
junior, senior, 1st year MS, etc):
Major:
Why are you taking this class?
What do you want to get out of the class?
What programming languages do you know?
23
End of Class Questions
24