Transcript Belief-optimal Reasoning for Cyber

CS B551: Elements of Artificial
Intelligence
Instructor: Kris Hauser
http://cs.indiana.edu/~hauserk
1
Basics

Class web site
• http://cs.indiana.edu/courses/b551

Textbook
• S. Russell and P. Norvig
• Artificial Intelligence: a Modern
Approach
• 2nd edition
2
Basics

Instructor
• Kris Hauser ([email protected])

AIs
• Ik Hyun Park ([email protected])
• Mark Wilson ([email protected])
3
Office Hours

Kris Hauser
• M,Th 1-2 in Lindley 301F

Ik Hyun Park
• Th 1:30-3:30 in TBA

Mark Wilson
• M 10-12 in Lindley 406
4
Agenda
Intro to AI
 Overview of class policies

5
Intro to AI
6
What is AI?

AI is the reproduction of human
reasoning and intelligent
behavior by computational methods
7
What is AI?

AI is an attempt of reproduction of
human reasoning and intelligent
behavior by computational methods
8
What is AI?

Discipline that systematizes and
automates reasoning processes to
create machines that:
Think like humans
Think rationally
Act like humans
Act rationally
9

Think like humans
Think rationally
Act like humans
Act rationally
The goal of AI is: to build machines that
operate in the same way that humans
think
• How do humans think?
• Build machines according to theory, test how
behavior matches mind’s behavior
• Cognitive Science


Manipulation of symbolic knowledge
How does hardware affect reasoning?
Discrete machines, analog minds
10


Think like humans
Think rationally
Act like humans
Act rationally
The goal of AI is: to build machines that perform
tasks that seem to require intelligence when
performed by humans
Take a task at which people are better, e.g.:
•
•
•
•
•


Prove a theorem
Play chess
Plan a surgical operation
Diagnose a disease
Navigate in a building
and build a computer system that does it
automatically
But do we want to duplicate human
imperfections?
11


Think like humans
Think rationally
Act like humans
Act rationally
The goal of AI is: to build machines that
make the “best” decisions given current
knowledge and resources
“Best” depending on some utility function
• Influences from economics, control theory


How do self-consciousness, hopes, fears,
compulsions, etc. impact intelligence?
Where do utilities come from?
12
What is Intelligence?
“If there were machines which bore a resemblance to
our bodies and imitated our actions as closely as
possible for all practical purposes, we should still
have two very certain means of recognizing that they
were not real men. The first is that they could never
use words, or put together signs, as we do in order
to declare our thoughts to others… Secondly, even
though some machines might do some things as well
as we do them, or perhaps even better, they would
inevitably fail in others, which would reveal that they
are acting not from understanding, …”
Discourse on the Method, by Descartes (1598-1650)
13
What is Intelligence?

Turing Test (c. 1950)
14
An Application of the Turing Test

CAPTCHA: Completely Automatic Public Turing
tests to tell Computers and Humans Apart
15
Chinese Room (John Searle)
16
Can Machines Act/Think Intelligently?

Yes, if intelligence is narrowly defined
as information processing
AI has made impressive achievements showing that
tasks initially assumed to require intelligence can be
automated
Each success of AI seems to push further the limits
of what we consider “intelligence”
17
Some Achievements




Computers have won over world
champions in several games, including
Checkers, Othello, and Chess, but still
do not do well in Go
AI techniques are used in many
systems: formal calculus, video games,
route planning, logistics planning,
pharmaceutical drug design, medical
diagnosis, hardware and software
trouble-shooting, speech
recognition, traffic monitoring,
facial recognition,
medical image analysis, part
inspection, etc...
DARPA Grand Challenge:
robotic car autonomously traversed
132 miles of desert
Some industries (automobile,
electronics) are highly robotized,
while other robots perform brain
and heart surgery, are rolling
on Mars, fly autonomously, …,
but home robots still remain
a thing of the future
18
18
Can Machines Act/Think Intelligently?

Yes, if intelligence is narrowly defined
as information processing
AI has made impressive achievements showing that
tasks initially assumed to require intelligence can be
automated

Maybe yes, maybe not, if intelligence
cannot be separated from
consciousness
 Is the machine experiencing thought?
 Strong vs. Weak AI
19
20
Big Open Questions

Is intelligent behavior just information
processing?
(Physical symbol system hypothesis)


If so, can the human brain solve problems
that are inherently intractable for
computers? Will a general theory of
intelligence emerge from neuroscience?
In a human being, where is the interface
between “intelligence” and the rest of
“human nature”
• Self-consciousness, emotions, compulsions

What is the role of the body?
(Mind-body problem)
21



AI contributes to building an information
processing model of human beings, just as
Biochemistry contributes to building a
model of human beings based on biomolecular interactions
Both try to explain how a human being
operates
Both also explore ways to avoid human
imperfections (in Biochemistry, by engineering
new proteins and drug molecules; in AI, by
designing rational reasoning methods)


Both try to produce new useful
technologies
Neither explains (yet?) the true meaning
of being human
22
Main Areas of AI







Knowledge
representation (including
formal logic)
Search, especially
heuristic search
(puzzles, games)
Planning
Reasoning under
uncertainty, including
probabilistic reasoning
Learning
Robotics and perception
Natural language
processing
Agent
Robotics
Reasoning
Search
Perception
Learning
Knowledge Constraint
rep.
satisfaction
Planning
Natural
language
...
Expert
Systems
23
Bits of History






1956: The name “Artificial Intelligence” is coined
60’s: Search and games, formal logic and
theorem proving
70’s: Robotics, perception, knowledge
representation, expert systems
80’s: More expert systems, AI becomes an
industry
90’s: Rational agents, probabilistic reasoning,
machine learning
00’s: Systems integrating many AI methods,
machine learning, reasoning under uncertainty,
robotics again
24
Syllabus

Introduction to AI
• Philosophy, history, agent frameworks

Search
• Uninformed search, heuristic search, heuristics

Search applications (and variants)
• Constraint satisfaction, planning, game playing, motion
planning

Reasoning under uncertainty
• Probability, planning under uncertainty, Bayesian
networks, probabilistic inference, dynamic modeling

Intro to machine learning
• Neural nets, decision tree learning, support vector
machines, etc.
25
Game theory
Computer Vision
E626
B657
B551
Biologically-inspired computing
B553
I486
Knowledge representation and
learning
B552
S626
S675
Robotics
B335
Q360
???
Topics in AI
Natural Language Processing
B659
B651
Q570
26
Careers in AI

‘Pure’ AI
• Academic, some labs

Applied AI
• Almost any area of CS!
• NLP, vision, robotics
• Economics

Cognitive Science
27
AI References

Conferences
• IJCAI, ECAI, AAAI, NIPS

Journals
• AI, Comp. I, IEEE Trans. Pattern Anal.
Mach. Intel., IEEE Int. Sys., Journal of
AIR

Societies
• AAAI, SIGART, AISB

AI Magazine (David Leake)
28
Class Policies
29
Grading

60% Homework
• Lowest score will be dropped
30% Final
 10% Participation

30
Programming Assignments
Projects will be written in Python
 Great for scripting

• Peter Norvig, Director of Research at
Google, and textbook author
Easy to learn
 2 weeks for each assignment

31
Homework Policy

Due at end of class on due date
• Typically Tuesdays

Extensions only granted in rare cases
• Require advance notice except
emergencies
32
Final Project

Encouraged if you are intending to do
research or coursework in AI, pursue
higher degree
• Individual or small groups (up to 3)
• Counts for 20% of homework grade

Content
• Software, new research, or technical report
• Mid-semester project proposal
• End-of-year report and in-class presentation
33
Enrollment

Add/drop deadline
• No penalty: Sept 4
• Late drop/add: Oct 28

Waitlist deadline: Sept 5
34
Swine Flu
35
Takeaways

AI has many interpretations
• Act vs. think, human-like vs. rational
• Concept has evolved

‘I’ has many interpretations
• Turing test
• Chinese room

AI success stories from each
perspective
36
Homework
Register
 Textbook
 Survey
 http://cs.indiana.edu/classes/b551
 Readings: R&N Ch. 1, 2, 26

37
What is Intelligence?

Total Turing Test
• Physical interaction
38