Download INTRODUCTION TO ARTIFICIAL INTELLIGENCE

Artificial Intelligence
CAP492
Dr. Souham Meshoul
Information Technology Department
CCIS – King Saud University
Riyadh, Saudi Arabia
[email protected]
CAP 492 Course
Dr. Souham Meshoul
1
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
Chapter 1
CAP 492 Course
Dr. Souham Meshoul
2
INTRODUCTION TO ARTIFICIAL INTELLIGENCE

Goal of Artificial Intelligence: Not only to understand how does mind work? but also how to build
intelligent entities?.
Engineering point of view: -Solve real-world problems using knowledge and reasoning
-Develop concepts, theory and practice of building intelligent entities
- Emphasis on system building
Scientific point of view:
- Use computers as a platform for studying intelligence itself
- Emphasis on understanding intelligent behavior.

Artificial Intelligence is one of the newest sciences which emerged after the world war II. AI represents a
big and open field.

The name Artificial Intelligence was adopted for the first time in 1956. (Computational Intelligence)

Artificial Intelligence can be viewed as a universal field: Ho to automate intellectual tasks?
CAP 492 Course
Dr. Souham Meshoul
3
INTRODUCTION TO ARTIFICIAL INTELLIGENCE

What is artificial Intelligence?
Several definitions are available in the literature.
Thinking vs Behavior
Model humans vs Work from an ideal standard

Two points of views:
1. Thinking/Acting humanly: success is measured in term of fidelity to human performance.
2. Thinking/Acting rationally: success is measured using an ideal concept of intelligence called
Rationality.

Rational System = system which does the « right thing » given what it knows.
CAP 492 Course
Dr. Souham Meshoul
4
INTRODUCTION TO ARTIFICIAL INTELLIGENCE

Some AI Definitions:

According to thought processes and reasoning
Thinking like humans:
« The exciting new effort to make computers think…machines with minds, in the full and literal sense.
» (Haugeland, 1985).
« The automation of activities that we associate with human thinking, activities such as decisionmaking, problem solving, learning… » (bellman, 1978).
Thinking rationally:
« The study of mental faculties through the use of computational models. » (Charniak and Mcdermott,
1985).
« The study of the computations that make it possible to perceive, reason and act. » (Winston 1992).
CAP 492 Course
Dr. Souham Meshoul
5
INTRODUCTION TO ARTIFICIAL INTELLIGENCE


Some AI Definitions:
According to behavior
Acting Humanly
« The art of creating machines that perform functions that require intelligence when performed by people. »
(Kurzweil, 1990)
« The study of how to make computers do things at which, at the moment, people are better. » (Rich and Knight,
1991)
Acting rationally
« Computational intelligence is the study of the design of intelligent agents. » (Poole et al.,1998).
« AI…is concerned with intelligent behavior in artifacts. » (Nilsson, 1998).
CAP 492 Course
Dr. Souham Meshoul
6
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 Some other definitions
•
Weak AI: (Searle) AI develops useful, powerful applications
•
Strong AI: claims machines have cognitive minds comparable to humans
debates on strong AI bring up some difficult conceptual problems in philosophy..
 Keep this one : Science of making machines do things that would require
intelligence if done by humans
Marvin Minsky
CAP 492 Course
Dr. Souham Meshoul
7
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 Acting Humanly
Turing Test: Alan Turing 1950 « Computing Machinery and Intelligence »
CAP 492 Course
Dr. Souham Meshoul
8
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
Test: The computer passes the “test of intelligence” if a human, after posing some written
questions, cannot tell whether the responses were from a person or not.
 Provides an operational definition of AI.
 Still being used today within the AI community in annual competitions – the Loebner Prize
($100,000 and a true 18K gold medal)

To give an answer, the computer would need to posses some capabilities:






Natural language processing: To communicate successfully.
Knowledge representation: To store what it knows or hears.
Automated reasoning: to answer questions and draw conclusions using stored information.
Machine learning: To adapt to new circumstances and to detect and extrapolate patterns.
Computer vision: To perceive objects.
Robotics to manipulate objects and move.
CAP 492 Course
Dr. Souham Meshoul
9
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
Example: ELIZA program, some extracts from the test:
 Patient: Men are all alike.
 ELIZA: In what way?
 Patient: They are always bugging about something or the other.
 ELIZA: Can you think of a specific example?
 Patient: Well, my boyfriend made me come here.
 ELIZA: Your boyfriend made you come here?
 Patient: He says I’m depressed much of the time.
 ELIZA: I am sorry you are depressed.
Problems:
 ELIZA and similar programs stressed simple syntactic analysis and generation of sentences.
They used pattern matching with known sentences.
 Turing’s test is not reproducible or amenable to mathematical analysis (it suggests duplicate
an examplar).
 An alternative: Study the underlying principles of intelligence (Wright brothers)
CAP 492 Course
Dr. Souham Meshoul
10
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 Thinking Humanly:
Program think like human → How humans think?
Requires Scientific theories of internal activities of the brain (cognitive science and cognitive
neuroscience).
Example:
 The General Problem Solver (GPS designed by Newell and Simon In 1963) was meant to be a
program that simulated human thought.

GPS: used means-end analysis in its search for solutions, computing the difference between
the goal and current, and then attempting to minimize the difference.

Newell and Simon by comparing GPS traces with those of human subjects discovered that
the behavior of GPS was largely a subset of human behavior
CAP 492 Course
Dr. Souham Meshoul
11
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 Thinking rationally:
The Laws of Thought approach is based on pattern for argument structure arising from
Aristostle’s syllogisms.
Example, “Socrates is a man; all men are mortal, therefore Socrates is mortal.” The laws of
thought initiated the field of logic.
The formal logic movement was advanced by Peano, Boole, Frege,, Godell and others (late
1800’s and early 1900’s)
Inspired perhaps by early progress, Hibert became a proponent of a school of thought known as
logicism or formalism. The goal of this was to devise a logic, or formal system, capable of
deriving all mathematical theorems.
CAP 492 Course
Dr. Souham Meshoul
12
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 Acting rationally:
 Modern AI can be characterized as the engineering of rational agents.
 An agent is simply an entity that perceives and acts. A rational agent is an entity
that perceives, reasons and acts rationally (correctly).
CAP 492 Course
Dr. Souham Meshoul
13
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 Foundations:
An interdisciplinary subject found on:







Philosophy,
mathematics,
economics,
neuroscience,
psychology,
computer engineering,
linguistics, and so on
CAP 492 Course
Dr. Souham Meshoul
14
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 History of Artificial Intelligence
Big dream
 Ultimately, we are dealing with the question: “What are we (human beings) doing
when we are thinking?”
 Thought processes in the human mind are computational in nature. There are
mechanistic procedures for generating these thoughts.
 Such computations can be simulated and implemented by a Turing machine.
Therefore, it can be programmed.
CAP 492 Course
Dr. Souham Meshoul
15
INTRODUCTION TO ARTIFICIAL INTELLIGENCE

History of Artificial Intelligence
Early days (1943-1955)

1943: first piece of AI work: Warren McCulloch and Walter Pitts

Model of artificial neurons

Mathematical learnable functions that generate “on/off” depending on inputs (logic gates)

Any computable function can be computed by a network of connected neurons.

Suitably defined networks can learn.

1949: Hebbian learning

A mechanism for updating the connection strength of a neuron.

Today, neurologists have confirmed that something similar to Hebbian learning indeed is going on in our brain
when we are learning.


1950: Turing test, complete vision of AI in “computing machinery and Intelligence”
1951: first neural network computer

Implemented by M. Minsky and D. Edmonds
CAP 492 Course
Dr. Souham Meshoul
16
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 History of Artificial Intelligence
Early days (1943-1955): Mcculloch and pitts artificial neuron
1
0
0.3
-1

1
0.5
CAP 492 Course
Dr. Souham Meshoul
17
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 History of Artificial Intelligence
Birth of AI 1956

1956: Dartmouth Conference



Organized by John McCarthy and colleagues for starting a new area in studying
computation and intelligence.
John McCarthy introduced the term “artificial intelligence” in the conference.
The next 20 years witnessed steady growth of the field led by the pioneers appeared in the
Dartmouth conference.
CAP 492 Course
Dr. Souham Meshoul
18
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 History of Artificial Intelligence
Expectations and Initial enthusiasm (1952 – 1969)
1956: Samuel’s checkers program:
First game playing program achieving human-competitive performance.
1957: Simon’s general problem solver (GPS):
Imitates the way a human would solve planning problems.
1958: Invention of LISP by J. McCarthy.
The first AI programming language.
1958: Minsky’s microworlds
The concept of creating a “controlled environment” in which problem solving
appears to require intelligence was born. The study of computation and
intelligence can become more manageable in these micro-worlds
CAP 492 Course
Dr. Souham Meshoul
19
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
Expectations and Initial enthusiasm (1952 – 1969)
1963: Thomas Evan’s program ANALOG
Solved analogy problems in an IQ test.
1965: ELIZA
Simulates a dialog with a computer in English on any topic.
Became popular when programmed to simulate a psychotherapist (Fedora’s
Emacs).
1967: Dendral program (developed at Stanford)
First successful program for scientific reasoning – one of the earlier rule based expert systems. A
program that can infer molecular structures given the information provided by a mass
spectrometer (that gives the masses of the various fragments of a molecule). The program relies on
expert knowledge (encoded as rules) to constraint the generation of possible molecular structures
that are consistent with the information from the mass spectrometer
CAP 492 Course
Dr. Souham Meshoul
20
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 History of Artificial Intelligence
Reality Check (1966 – 1973): series of disappointments and frustrations
AI was poured little buckets of “reality cold water”
Problems
 Most early systems contain little or no knowledge of their subject matter



Knowledge acquisition bottleneck.
Example: Poor performance of earlier machine translation system (Russian  English): “the spirit is willing
but the flesh is weak” was translated to “the vodka is good but the meat is rotten”.
Computational Intractability of AI problems
 Theory of computational complexity was not developed. Polynomial solvable problems, NP-completeness, etc
 People thought a faster machine could solve any hard problem.
 Initial frustration with theorem proving led to a disappointment in AI. Theorem proving is exponential in
complexity
CAP 492 Course
Dr. Souham Meshoul
21
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 History of Artificial Intelligence
Resurgence (1969 – 1979)

1971: T. Winograd’s Ph.D. thesis (MIT) demonstrated a system that can understand English in
a micro-domain (the block world).

1972: PROLOG was developed by a group of Europeans and became alternative to LISP as an
AI programming language.

1974: MYCIN was developed by Ted Shortliffe. Expert system for medical diagnosis.
Sometimes called the first expert system.

1978: The Version Space algorithm was developed by Tom Mitchell at Stanford. First
symbolic machine learning algorithm. “Father of Machine Learning”.
 1979: Non-monotonic logic. Began to be formalized by John McCarthy and his colleagues.
CAP 492 Course
Dr. Souham Meshoul
22
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 History of Artificial Intelligence
Resurgence (1969 – 1979): Winograd 1972
CAP 492 Course
Dr. Souham Meshoul
23
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 History of Artificial Intelligence
AI becomes an industry (1980 – present)
 AI started to become industrially and commercially beneficial
1982: R1 was deployed at DEC – an expert system that saved the company around
$40M / year
Du Pont had 100 in use and an estimated 500 in development at late 90’s to early 21st
century
 At an international level, AI was considered a part of a country’s technological
developments
Japan: “First Generation” project (10 year plan to build intelligence machines running
in Prolog)
USA: Microelectronics and Computer Technology Corporation (MCC) was formed in
response
Britain: Funding for AI was reinstated
CAP 492 Course
Dr. Souham Meshoul
24
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 History of Artificial Intelligence
Renewing with connectionism and AI becomes a science (1986 – present)
 Work of the physicist John Hopfield (1982) on using techniques from statistical
mechanics.
 Connectionist models of intelligent systems competitor to the symbolic models
(Newell and Simon) and logicist approach (McCarthy). (complementary approaches
in fact).
 Several revolutions in many fields: pattern recognition, computer vision, robotics…
 Emergence of intelligent agents.
CAP 492 Course
Dr. Souham Meshoul
25
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
Examples of AI applications: Game Playing
 TDGammon, the world champion backgammon player, built by Gerry Tesauro of IBM
research.
Perception: keyboard input.
Reason: reinforcement learning.
Actuation: graphical output shows dice and movement of piece.
 Deep Blue chess program beat world champion Gary Kasparov
Perception: input symptoms and test results.
Reason: Bayesian networks, Monte-Carlo simulations.
Actuation: output diagnoses and further test suggestions.
CAP 492 Course
Dr. Souham Meshoul
26
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 Examples of AI applications: Natural Language Understanding
 Natural language understanding (spell checkers, grammar checkers)
 AI translators – spoken to and prints what one wants in foreign languages : Alta
Vista’s translation of web pages.
 Advanced systems can answer questions based on the information in the text
and produce useful summaries.
 PROVERB (Littman 1999) crossword puzzles
 Examples of successes : English conversation
START system: accesses raw data tables, and then can carry on a dialogue
CAP 492 Course
Dr. Souham Meshoul
27
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
Examples of AI applications: Expert systems

In geology
prospector expert system carries evaluation of mineral potential of geological site
or region

Diagnostic Systems
Pathfinder, a medical diagnosis system (suggests tests and makes diagnosis)
developed by Heckerman and other Microsoft research
Microsoft Office Assistant in Office provides customized help by decisiontheoretic reasoning by an individual user.
MYCIN system for diagnosing bacterial infections of the blood and suggesting
treatments

System Configuration
"XCON" (for custom hardware configuration) configures computers doing work
of 300 people using 10,000 rules
CAP 492 Course
Dr. Souham Meshoul
28
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 Examples of AI applications: Robotics
 Robotics becoming increasing important in various areas like: games, to handle
hazardous conditions and to do tedious jobs among other things.
 Examples: automated cars, ping pong player, mining, construction, robot assistant
in microsurgery,…
CAP 492 Course
Dr. Souham Meshoul
29
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 Main issues in AI


Representation
Search: many tasks can be viewed as searching a very large problem space for
solution space

Inference: related to search, inferring other facts from some given facts. e.g.,

Learning: inductive inference, neural networks, artificial life, genetic algorithms,
knowing all “elephants have trunks” and “Jo is an elephant,” can we answer does
Jo have a trunk?
evolutionary strategies

Planning: starting with general facts about the world, facts about the effects of basic
actions, facts about a particular situation, and a statement of a goal, generate a
strategy for achieving that goal in terms of a sequence of primitive steps or actions
CAP 492 Course
Dr. Souham Meshoul
30
INTRODUCTION TO ARTIFICIAL INTELLIGENCE
 Summary
 Intelligence is studied from many perspectives: Are you concerned with thinking or
behavior?
 AI can help us solve difficult, real-world problems, creating new opportunities in
business, engineering, and many other application areas.
 The history of AI has had cycles of success, misplaced optimism, and resulting
cutbacks in enthusiasm and funding. There have also been cycles of introducing
new creative approaches and systematically refining the best ones.
 AI has advanced more rapidly in the past decade because of greater use of the
scientific method in experimenting with and comparing approaches.
CAP 492 Course
Dr. Souham Meshoul
31