Transcript Artificial Intelligence Overview

Artificial Intelligence Overview
John Paxton
Montana State University
February 22, 2005
[email protected]
Montana State University
A Brief Bio
• 1985: The Ohio State University, B.S.
• 1987: The University of Michigan, M.S.
• 1990: The University of Michigan, Ph.D.
• 1990 – present: MSU CS Professor
Talk Outline
•
•
•
•
•
•
•
What is AI?
Foundations
Areas
Search
Knowledge Representation
Agents
Questions
What is AI?
Scientific Approach
1. Build systems that think like humans
2. Build systems that act like humans
Engineering Approach
1. Build systems that think rationally
2. Build systems that act rationally
Acting Like a Human
• Turing Test (1950)
IBM
Thinking Like a Human
• Cognitive Modeling Approach
• General Problem Solver (Newell and
Simon, 1961)
• Towers of Hanoi Problem
Thinking Rationally
• The laws-of-thought approach
• Syllogisms (Aristotle): deductive reasoning
in which a conclusion is derived from
premises
• It is difficult to code the knowledge and to
reason with it efficiently.
Sample Logic Puzzle
• Robinson found himself on an island where some of the
people were liars, and others always told the truth. When
he met with one of the inhabitant of the island, he asked
him: "Are you a liar or not?"
"I'm not a liar", answered the person.
"All right, if it is so, you'll be my companion", Robinson
said.
After a while they saw another man.
Robinson pointed to the man and asked his new friend,
"Could you, please, ask him, if he is a liar or not?"
The new friend asked the question to the man, came
back and said,
"He said he was not a liar".
"All right, now I'm convinced that you are not a liar!"
smiled Robinson. What convinced Robinson?
Acting Rationally
• Rational Agent Approach. The agent acts
to achieve the best (or near best)
expected outcome.
Water Jug Problem
4
5
10
2
15
8
10
4
5
Foundations
•
•
•
•
•
•
•
•
Philosophy (e.g. Where does knowledge come from?)
Mathematics (e.g. What are the formal rules to draw valid conclusions?)
Economics (e.g. How should we make decisions to maximize payoff?)
Neuroscience (e.g. How do brains process information?)
Psychology (e.g. How do humans and animals think and act?)
Computer Engineering (e.g. How can we build an efficient computer?)
Control Theory (e.g. How can artifacts operate under their own control?)
Linguistics (e.g. How does language relate to thought?)
Areas
•
•
•
•
•
•
•
Agents
Artificial Life
Machine Discovery and Data Mining
Expert Systems
Fuzzy Logic
Game Playing
Genetic Algorithms
Areas
•
•
•
•
•
•
•
Knowledge Representation
Learning
Neural Networks
Natural Language Processing
Planning
Reasoning
Robotics
Areas
•
•
•
•
Search
Speech Recognition and Synthesis
Virtual Reality
Computer Vision
Search
• Missionaries and Cannibals Problem
MMM
CCC
Search
• Missionaries and Cannibals Solution
MMM
CCC
MMM
C
CC
MM
CC
M
C
M
C
MM
CC
MMM
CC
MM
CC
C
M
C
MMM
CC
CCC
MMM
C
Types of Search
• Uninformed Search
– Breadth-First Search
– Depth-First Search
• Informed Search
– Best-First Search
– A* Search
Breadth-First Search
MMM
CCC
MMM
CC
C
MMM
C
CC
MM
CC
M
C
Knowledge Representation
• Semantic Nets
• Fuzzy Logic
• First Order Predicate Calculus
Supply the Missing Words!
•
•
•
•
•
60 = M in an H
26 = L in the A
12 = S of the Z
88 = P K
200 = D for P G in M
Semantic Nets
can-fly
yes
bird
is-a
is-a
is-a
robin
magpie
no
ostrich
can-fly
Fuzzy Logic
• Shaquille O’Neal is tall
1.0
tall
0.0
5’0
6’0
7’0
First Order Predicate Calculus
• Every Saturday is a weekend.
x Saturday(x)  weekend(x)
• Some day is a week day.
x day(x)  weekday(x)
Agents
sensors
actuators
AGENT
ENVIRONMENT
Rationality Factors
• Performance Measure
• Prior Knowledge
• Performable Actions
• Agent’s Prior Percepts
Rational Agent
• For each possible sensor sequence, a
rational agent should select an action that
is expected to maximize its performance
measure, given the evidence provided by
the sensor sequence and whatever built-in
knowledge the agent has.
Thank you!
• Questions??