intro-class.ppt

Intorduction to
Artificial Intelligence
Rina Dechter
CS 171
Fall 2006

271- Fall 2006
Robotic links
 Robocup Video
 Soccer Robocupf
 Darpa Challenge
 Darpa’s-challenge-video
 http://www.darpa.mil/grandchallenge05/TechPapers/Stanford.pdf

271- Fall 2006
CS171
 Course home page:
http://www.ics.uci.edu/~dechter/ics-171/fall-06/
 schedule, lecture notes, tutorials, assignment, grading,
office hours, etc.


 Textbook: S. Russell and P. Norvig Artificial
Intelligence: A Modern Approach Prentice Hall, 2003,
Second Edition
 Grading: Homeworks and projects (30-40%)
 Midterm and final (60-70%)



271- Fall 2006
Course overview
 Introduction and Agents (chapters 1,2)
 Search (chapters 3,4)
 Games (chapter 5)
 Constraints processing (chapter 6)
 Representation and Reasoning with Logic
(chapters 7,8,9)
 Learning (chapters 18,20)
 Planning (chapter 11)
 Uncertainty (chapters 13,14)
 Natural Language Processing (chapter 22,23)

271- Fall 2006
Course Outline
Resources on the Internet
 AI on the Web: A very comprehensive list of
Web resources about AI from the Russell and
Norvig textbook.
Essays and Papers
 What is AI, John McCarthy
 Computing Machinery and Intelligence, A.M.
Turing
 Rethinking Artificial Intelligence, Patrick
H.Winston

271- Fall 2006
Today’s class
 What is Artificial Intelligence?
 A brief History
 Intelligent agents
 State of the art

271- Fall 2006
What is Artificial Intelligence
(John McCarthy , Basic Questions)
 What is artificial intelligence?
 It is the science and engineering of making intelligent machines, especially
intelligent computer programs. It is related to the similar task of using
computers to understand human intelligence, but AI does not have to confine
itself to methods that are biologically observable.
 Yes, but what is intelligence?
 Intelligence is the computational part of the ability to achieve goals in the
world. Varying kinds and degrees of intelligence occur in people, many
animals and some machines.
 Isn't there a solid definition of intelligence that doesn't depend on
relating it to human intelligence?
 Not yet. The problem is that we cannot yet characterize in general what kinds
of computational procedures we want to call intelligent. We understand some
of the mechanisms of intelligence and not others.
 More in: http://www-formal.stanford.edu/jmc/whatisai/node1.html

271- Fall 2006
What is AI?
Views of AI fall into four categories:
Thinking humanly Thinking rationally
Acting humanly Acting rationally
The textbook advocates "acting rationally“
List of AI-topics

271- Fall 2006
What is Artificial
Intelligence?
 Human-like (“How to simulate humans intellect and
behavior on by a machine.)
 Mathematical problems (puzzles, games, theorems)
 Common-sense reasoning (if there is parking-space,
probably illegal to park)
 Expert knowledge: lawyers, medicine, diagnosis
 Social behavior
 Rational-like:
 achieve goals, have performance measure

271- Fall 2006
What is Artificial
Intelligence
 Thought processes
 “The exciting new effort to make computers
think .. Machines with minds, in the full and
literal sense” (Haugeland, 1985)
 Behavior
 “The study of how to make computers do
things at which, at the moment, people are
better.” (Rich, and Knight, 1991)

271- Fall 2006
The Turing Test
(Can Machine think? A. M. Turing, 1950)
 Requires
 Natural language
 Knowledge representation
 Automated reasoning
 Machine learning
 (vision, robotics) for full test

271- Fall 2006
What is AI?
 Turing test (1950)
 Requires:
 Natural language
 Knowledge representation
 automated reasoning
 machine learning
 (vision, robotics.) for full test
 Thinking humanly:
 Introspection, the general problem solver (Newell and
Simon 1961)
 Cognitive sciences
 Thinking rationally:
 Logic
 Problems: how to represent and reason in a domain
 Acting rationally:
 Agents: Perceive and act

271- Fall 2006
AI examples
Common sense reasoning
 Tweety
 Yale Shooting problem
Update vs revise knowledge
 The OR gate example: A or B - C
 Observe C=0, vs Do C=0
Chaining theories of actions
Looks-like(P)  is(P)
Make-looks-like(P)  Looks-like(P)
----------------------------------------
Makes-looks-like(P) ---is(P) ???
Garage-door example: garage door not included.
 Planning benchmarks
 8-puzzle, 8-queen, block world, grid-space world
Abduction: cambridge parking example

271- Fall 2006
History of AI
 McCulloch and Pitts (1943)
 Neural networks that learn
 Minsky (1951)
 Built a neural net computer
 Darmouth conference (1956):
 McCarthy, Minsky, Newell, Simon met,
 Logic theorist (LT)- proves a theorem in Principia
Mathematica-Russel.
 The name “Artficial Intelligence” was coined.
 1952-1969
 GPS- Newell and Simon
 Geometry theorem prover - Gelernter (1959)
 Samuel Checkers that learns (1952)
 McCarthy - Lisp (1958), Advice Taker, Robinson’s
resolution
 Microworlds: Integration, block-worlds.
 1962- the perceptron convergence (Rosenblatt)

271- Fall 2006
The Birthplace of
“Artificial Intelligence”, 1956
 Darmouth workshop, 1956: historical meeting of the precieved
founders of AI met: John McCarthy, Marvin Minsky, Alan
Newell, and Herbert Simon.
 A Proposal for the Dartmouth Summer Research Project on
Artificial Intelligence. J. McCarthy, M. L. Minsky, N.
Rochester, and C.E. Shannon. August 31, 1955. "We propose
that a 2 month, 10 man study of artificial intelligence be
carried out during the summer of 1956 at Dartmouth College
in Hanover, New Hampshire. The study is to proceed on the
basis of the conjecture that every aspect of learning or any
other feature of intelligence can in principle be so precisely
described that a machine can be made to simulate it." And this
marks the debut of the term "artificial intelligence.“
 50 anniversery of Darmouth workshop

271- Fall 2006
History, continued
 1966-1974 a dose of reality
 Problems with computation
 1969-1979 Knowledge-based systems
 Weak vs. strong methods
 Expert systems:
• Dendral:Inferring molecular structures
• Mycin: diagnosing blood infections
• Prospector: recomending exploratory drilling (Duda).
 Roger Shank: no syntax only semantics
 1980-1988: AI becomes an industry
 R1: Mcdermott, 1982, order configurations of computer
systems
 1981: Fifth generation
 1986-present: return to neural networks
 Recent event:
 AI becomes a science: HMMs, planning, belief network

271- Fall 2006
Abridged history of AI
 1943 McCulloch & Pitts: Boolean circuit model of brain
 1950 Turing's "Computing Machinery and Intelligence"
 1956 Dartmouth meeting: "Artificial Intelligence" adopted
 1952—69 Look, Ma, no hands!
 1950s Early AI programs, including Samuel's checkers
program, Newell & Simon's Logic Theorist,
Gelernter's Geometry Engine
 1965 Robinson's complete algorithm for logical reasoning
 1966—73 AI discovers computational complexity
Neural network research almost disappears
 1969—79 Early development of knowledge-based systems
 1980-- AI becomes an industry
 1986-- Neural networks return to popularity
 1987-- AI becomes a science
 1995-- The emergence of intelligent agents

271- Fall 2006
State of the art
 Deep Blue defeated the reigning world chess
champion Garry Kasparov in 1997
 Proved a mathematical conjecture (Robbins
conjecture) unsolved for decades
 No hands across America (driving autonomously 98%
of the time from Pittsburgh to San Diego)
 During the 1991 Gulf War, US forces deployed an AI
logistics planning and scheduling program that
involved up to 50,000 vehicles, cargo, and people
 NASA's on-board autonomous planning program
controlled the scheduling of operations for a spacecraft
 Proverb solves crossword puzzles better than most
humans
 DARPA grand challenge 2003-2005, Robocup

271- Fall 2006
Robotic links
Robocup Video
 Soccer Robocupf
Darpa Challenge
 Darpa’s-challenge-video

271- Fall 2006
Agents (chapter 2)
 Agents and environments
 Rationality
 PEAS (Performance measure,
Environment, Actuators, Sensors)
 Environment types
 Agent types

271- Fall 2006
Agents
 An agent is anything that can be viewed as
perceiving its environment through sensors
and acting upon that environment through
actuators

 Human agent: eyes, ears, and other organs
for sensors; hands,
 legs, mouth, and other body parts for
actuators

 Robotic agent: cameras and infrared range
finders for sensors;

271- Fall 2006
Agents and environments
 The agent function maps from percept
histories to actions:

[f: P*  A]
 The agent program runs on the physical
architecture to produce f

271- Fall 2006
Vacuum-cleaner world
 Percepts: location and contents, e.g.,
[A,Dirty]

 Actions: Left, Right, Suck, NoOp


271- Fall 2006
Rational agents
 An agent should strive to "do the right
thing", based on what it can perceive and
the actions it can perform. The right action
is the one that will cause the agent to be
most successful

 Performance measure: An objective
criterion for success of an agent's behavior

 E.g., performance measure of a vacuum-
cleaner agent could be amount of dirt
cleaned up, amount of time taken, amount
of electricity consumed, amount of noise

271- Fall 2006
Rational agents
 Rational Agent: For each possible
percept sequence, a rational agent
should select an action that is
expected to maximize its performance
measure, given the evidence provided
by the percept sequence and
whatever built-in knowledge the agent
has.


271- Fall 2006
What’s involved in Intelligence?
Intelligent agents
 Ability to interact with the real world
 to perceive, understand, and act
 e.g., speech recognition and understanding and
synthesis
 e.g., image understanding
 e.g., ability to take actions, have an effect
 Knowledge Representation, Reasoning and
Planning
 modeling the external world, given input
 solving new problems, planning and making decisions
 ability to deal with unexpected problems, uncertainties
 Learning and Adaptation
 we are continuously learning and adapting
 our internal models are always being “updated”
• e.g. a baby learning to categorize and recognize
animals

271- Fall 2006
Implementing agents
 Table look-ups
 Autonomy
 All actions are completely specified
 no need in sensing, no autonomy
 example: Monkey and the banana
 Structure of an agent
 agent = architecture + program
 Agent types
• medical diagnosis
• Satellite image analysis system
• part-picking robot
• Interactive English tutor
• cooking agent
• taxi driver

271- Fall 2006
Agent types
 Example: Taxi driver
 Simple reflex
 If car-in-front-is-breaking then initiate-breaking
 Agents that keep track of the world
 If car-in-front-is-breaking and on fwy then initiate-
breaking
 needs internal state
 goal-based
 If car-in-front-is-breaking and needs to get to hospital
then go to adjacent lane and plan
 search and planning
 utility-based
 If car-in-front-is-breaking and on fwy and needs to
get to hospital alive then search of a way to get to the
hospital that will make your passengers happy.
 Needs utility function that map a state to a real
function (am I happy?)

271- Fall 2006
Summary
 What is Artificial Intelligence?
 modeling humans thinking, acting, should think,
should act.
 History of AI
 Intelligent agents
 We want to build agents that act rationally
 Real-World Applications of AI
 AI is alive and well in various “every day” applications
• many products, systems, have AI components
 Assigned Reading
 Chapters 1 and 2 in the text R&N

intro-class.ppt

More Related Content

Similar to intro-class.ppt (19)

Recently uploaded (20)

intro-class.ppt