ARTIFICIAL

INTELLIGENCE: INTRODUCTION
 Short presentation
2

Prof. Dr. Hachemi BENNACEUR

‫ هاشمي بن ناصر‬.‫ د‬.‫أ‬
[email protected] : Email

:Exam
 20% Project
 20% Mid-term test
 20% Quizzes
 40% Final exam Textbook

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Grading
3

Attendance is important in this course.

Failure to attend 80% may result in denial
of your entry to the final exam, and a 0
for the course.

Data Structures ١٤٤٣/٠٣/١٥

 Course overview
4

 What is AI : Introduction and Intelligent Agents (chapters 1,2)

 Solving problems by searching (chapter 3)
 Informed search methods (chapter 4)
 Game playing (chapter 6)
 Knowledge and reasoning (chapter 7)
 Acting Logically, Prolog Language (chapters 8, 9)
 Constraint satisfaction problems (chapter 5)
 Learning (chapters 18,20)
 Present & Future AI

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Chapter1 : Outline
5

 What is AI
 A brief history
 The state of the art

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 … What is AI? Found on the Web
6

 AI is the simulation of intelligent human processes

 AI is the reproduction of the methods or results of
human reasoning or intuition
 AI is the study of mental faculties through the use
computational methods
 Using computational models to simulate intelligent
behavior
 Machines to emulate humans

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Intelligent
behavior
Computer

Humans

7 Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

8
?Why AI
 Cognitive Science: As a way to understand how natural
minds and mental phenomena work
 e.g., visual perception, memory, learning, language, etc.
 Philosophy: As a way to explore some basic and interesting
(and important) philosophical questions
 e.g., the mind body problem, what is consciousness, etc.
 Engineering: To get machines to do a wider variety of useful
things
 e.g., understand spoken natural language, recognize individual people
in visual scenes, find the best travel plan for your vacation, etc.

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Weak vs. Strong AI
9

 Weak AI: Machines can be made to behave as if they were

intelligent

 Strong AI: Machines can have consciousness

 Subject of fierce debate among philosophers and AI

researchers.

 E.g. Red Herring article and responses

 http://groups.yahoo.com/group/webir/message/1002

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 AI Characterizations
10

Discipline that systematizes and automates

intellectual tasks to create machines that :

Act like humans Act rationally

System passing the Turing Test (1950) Rational agent (199X)
Learning from Knowledge (adapt) acts according to his beliefs
Representing Knowledge (memorize) to reach goals
Solve Pb (argue) (not only logical)
Understanding (communicate) Pragmatic
Theoretical

Think like humans Think rationally

Cognitive modeling logical thinking
(GPS (Newel & Simon,61)) Pascal [1623-1662] (calculating machine)
Leibniz [1646-1716] (reasoning machine)
Complex Babbage [1792-1871] (Analytical Engine)
limited

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Systems that act like humans
11

 When does a system behave intelligently?

 Turing (1950) Computing Machinery and Intelligence
 "Can machines think?"  "Can machines behave intelligently?"
 Operational test of intelligence: imitation games
Interrogator interacts with a computer and
a person.

Computer passes the Turing test if

interrogator cannot determine which is
which.

 Test requires the collaboration of major components of AI: knowledge,

reasoning, language understanding, learning, …

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Systems that act like humans
12

 AI is the art of creating machines that perform

functions that require intelligence when
performed by humans
 Methodology: Take an intellectual task at which
people are better and make a computer do it
 Turing test •Prove a theorem
•Play chess
•Plan a surgical operation
•Diagnose a disease
•Navigate in a building
Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥
 Chess
13

Name: Garry Kasparov

Title: World Chess
Champion

Humans are still better at making up excuses.

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥
© Jonathan Schaeffer
 Perspective on Chess
14

“Saying Deep Blue doesn’t really think

about chess is like saying an airplane
doesn't really fly because it doesn't flap
its wings”
Drew McDermott

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

© Jonathan Schaeffer
 Systems that think like humans
15

 How do humans think?

 Requires scientific theories of internal brain activities (cognitive model):
 How to validate? requires :
 Predicting and testing human behavior
 Identification from neurological data
 Cognitive Science vs. Cognitive neuroscience.
 Both approaches are now distinct from AI
 Share that the available theories do not explain anything
resembling human intelligence.
 Three fields share a principal direction.

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Systems that think like humans
16

 Some references;
 Daniel C. Dennet.
Consciousness explained.
 M. Posner (edt.) Foundations
of cognitive science
 Francisco J. Varela et al. The
Embodied Mind
 J.-P. Dupuy. The
mechanization of the mind

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Systems that think rationally
17

 Capturing the laws of thought

 Aristotle: What are ‘correct’ argument and thought
processes?
 Correctness depends on irrefutability of reasoning processes.
 This study initiated the field of logic.
 The logicist tradition in AI hopes to create intelligent systems using
logic programming.
 Problems:
 Not all intelligence is expressed by logic behavior
 What is the purpose of thinking? What thought should one have?

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Systems that think rationally
18

 A reference;
 Ivan Bratko, Prolog
programming for
artificial intelligence.

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Systems that act rationally
19

 Rational behavior: “doing the right thing”

 The “Right thing” is that what is expected to maximize
goal achievement given the available information.
 Can include thinking, yet in service of
rational action.
 Action without thinking: e.g. reflexes.

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Systems that act rationally
20

Some references 

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Systems that act rationally
21

 Two advantages over previous approaches:

 More general than law of thoughts approach
 More amenable to scientific development.

 Yet rationality is only applicable in ideal environments.

 Moreover rationality is not a very good model of

reality.

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Think/Act Rationally
22

 Always make the best decision given what is

available (knowledge, time, resources)
 Perfect knowledge, unlimited resources  logical
reasoning
 Imperfect knowledge, limited resources 
(limited) rationality
•Connection to economics, operational research,
and control theory
•But ignores role of consciousness, emotions,
fear of dying on intelligence
Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥
 Rational agents
23

 An agent is an entity that perceives and acts

 This course is about designing rational agents
 An agent is a function from percept histories to actions:

f : P*  A
 For any given class of environments and task we seek the agent (or
class of agents) with the best performance.
 Problem: computational limitations make perfect rationality
unachievable.

 Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Problematic
24

The preferred areas of AI are :

-when there is tractability p,
as combinatorial problems : crypto-arithmetic, games,
crosswords, Planning, games, economics, ...
example: SEND + MORE = MONEY 9 x 8 x 7 x 6 x 5 x 4 x 3 x
2 = 362 880 possible trial.
- problems requiring heuristics.
example: chess (10160)
Heuristics are pragmatic knowledge expressing a know-
how, an expertise rather than a systematic calculation.
The AI is also designed to simulate human reasoning.
Modeling knowledge and reasoning of a human expert, and
making them accessible to non computer scientist.

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Foundations of AI
25

 Different fields have contributed to AI in the form of

ideas, view points and techniques.
 Philosophy: Logic, reasoning, mind as a physical system, foundations of learning,
language and rationality.
 Mathematics: Formal representation and proof algorithms, computation,
(un)decidability, (in)tractability, probability.
 Psychology: adaptation, phenomena of perception and motor control.
 Economics: formal theory of rational decisions, game theory.
 Linguistics: knowledge representation, grammar.
 Neuroscience: physical substrate for mental activities.
 Control theory: homeostatic systems, stability, optimal agent design.

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 History of AI
26

 Jean-Claude Latombe:
I personally think that AI is (was?) a rebellion
against some form of establishment telling
us “Computers cannot perform certain tasks
requiring intelligence”

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 A brief history
27

 What happened after WWII?

 1943: Warren Mc Culloch and Walter Pitts: a model of artificial
boolean neurons to perform computations.
 First steps toward connectionist computation and learning (Hebbian learning).
 Marvin Minsky and Dann Edmonds (1951) constructed the first neural network
computer

 1950: Alan Turing’s “Computing Machinery and Intelligence”

 First complete vision of AI.
 Idea of Genetic Algorithms

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 A brief history (2)
28

 The birth of (the term) AI (1956)

 Darmouth Workshop bringing together top minds on automata theory,
neural nets and the study of intelligence.
 Allen Newell and Herbert Simon: The logic theorist (first non-numerical thinking program
used for theorem proving).
 For the next 20 years the field was dominated by these participants.

 Great expectations (1952-1969)

 Newell and Simon introduced the General Problem Solver.
 Imitation of human problem-solving
 Arthur Samuel (1952-) investigated game playing (checkers ) with great success.
 John McCarthy(1958-) :
 Inventor of Lisp (second-oldest high-level language)
 Logic oriented, Advice Taker (separation between knowledge and reasoning)

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 A brief history (3)
29

 The birth of AI (1956)

 Great expectations continued ..
 Marvin Minsky (1958 -)
 Introduction of microworlds that appear to require intelligence to solve: e.g. blocks-world.
 Anti-logic orientation, society of the mind.
 Herbert Gelernter (1959) : constructed the geometry theorem Prover.
 Artur Samual (1952-1956) : a series of programs for checkers.
 Collapse in AI research (1966 - 1973)
 Progress was slower than expected.
 Unrealistic predictions.
 Some systems lacked scalability.
 Combinatorial explosion in search.
 Fundamental limitations on techniques and representations.

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 A brief history (4)
30

 AI revival through knowledge-based systems

(1969-1970)
 General-purpose vs. domain specific
 E.g. the DENDRAL chemistry project (Buchanan et al. 1969)
 First successful knowledge intensive system.
 Expert systems
 MYCIN to diagnose blood infections (Feigenbaum et al.)
 Introduction of uncertainty in reasoning.
 Increase in knowledge representation research.
 Logic, frames, semantic nets, …

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 A brief history (5)
31

 AI becomes an industry (1980 - present)

 First successful commercial expert system R1 at DEC (McDermott, 1982)
 Fifth generation project in Japan (1981) : a 10-year plan to build intelligent computer
running Prolog.
 American response …: US formed the microelectronics and Computer Technology
Corporation designed to assure national competitiveness (chip design and human-interface
research)

 Puts an end to the AI winter.

AI industry boomed from a few million to billion dollars in 1988. Period called “AI winter” in
which many companies suffered as they failed to deliver on extravagant promises.
promises

 Connectionist revival (1986 - present)

 Parallel distributed processing (RumelHart and McClelland, 1986);
back-propagation learning (computer science and psychology).

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 A brief history (6)
32

 AI becomes a science (1987 - present)

 In speech recognition: hidden markov models
 In neural networks
 In uncertain reasoning and expert systems: Bayesian network formalism
 Problem solving
 …

 The emergence of intelligent agents (1995 - present)

 The whole agent problem:
“How does an agent act/behave embedded in real environments with continuous sensory
inputs”
“Ideally, an intelligent agent takes the best possible action in a situation : study the
problem of building intelligent agents in this sense”.

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 State of the art : AI today 1/2

33

 Autonomous planning and scheduling : on-board autonomous planning

program to control the scheduling of operations for a spacecraft (Jonhson
et al., 2000).
 Game playing : IBM’s Deep Blue became the first computer program to
defeat the world champion in chess match (Goodman and Keene, 1997),
 Autonomous control : the ALVINN computer vision system was trained to
steer a car to keep it following a lane (for 2850 miles ALVINN was in
control of steering in 98%, only 2% for human control mostly at exit
ramps).
 Diagnosis : medical diagnosis programs based on probabilistic analysis
have been able to perform at level of an expert physician in several areas
in medicine (Heckerman 1991).

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 State of the art : AI today 2/2

34

 Logistics planning
 Robotics
 Language understanding and problem solving.

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Major research areas (Applications)
35

 Natural Language Understanding

 Image, Speech and pattern recognition

 Uncertainty Modeling

 Problem solving

 Knowledge representation

 …..

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Symbolic Programming
36

 Program as Representation of world

 Symbol as basic element of representation
 atom, property, relationship
 Symbolic Expression as method of combination
 LISP for Symbolic programming
 PROLOG for logic programming
 Object-Oriented Concept

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Knowledge Representation
37

 What kind of Knowledge needed for Problem

solving ?
 Structure of knowledge ?
 declarative vs procedural
 Representation techniques ?
 explicit vs (implicit + inference)
 logic, frame, object-oriented, semantic net, script
 Knowledge acquisition and update

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Search Theory
38

 An Optimization method
 Analyze alternative cases and select one
 Cope with Exponential complexity, NP classes
 Try likely one first (Heuristic Search)
 Utilize local information (Hill Climbing Method)
 Optimal solution vs good solution
 Genetic Algorithm, Simulated Annealing
 Stochastic search

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Genetic Algorithms
39

 Computational model of life evolution

 Stochastic optimization technique
 Initial chromosome creation
 New generations are made (crossover, mutation)
 survival of the fittest
 Base of artificial life research
 study evolution of life, by simulation

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 AI Success Story : Medical expert systems
40

Programs listed by Special Field

Antibiotics & Infectious    Gynecology
Diseases
  Imaging Analysis
Cancer 

Chest pain    Internal Medicine

Dentistry    Intensive Care
Dermatology 
  Laboratory Systems
Drugs & 
Toxicology   Orthopedics
Emergency    Pediatrics
Epilepsy 
  Pulmonology & Ventilation
Family Practice 
  Surgery & Post-Operative
Genetics 
Care
Geriatrics 
  Trauma Management
Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥
 Pattern Recognition Applications
41

 Handwriting and document recognition

 Signature, biometrics (finger, face, iris, etc.)

 Trafic monitoring, Remote Sensing

 guided missile, target homing

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Handwriting Understanding
42

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

전자 펜으로 수식 입력 수식 인식
 Future of AI
43

 Making AI Easy to use

 Easy-to-use Expert system building tools
 Web auto translation system
 Recognition-based Interface Packages
 Integrated Paradigm
 Symbolic Processing + Neural Processing
 AI in everywhere, AI in nowhere
 AI embedded in all products
 Ubiquitous Computing, Pervasive Computing

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Some references
44

 Understanding Intelligence by Rolf

Pfeifer and Christian Scheier.

 Artificial Intelligence: Structures and

Strategies for Complex Problem-solving
by George Luger.

 Computation and Intelligence: Collective

readings edited by George Luger.

 Paradigms of Artificial Intelligence

Programming: Case Studies in Common
Lisp by Peter Norvig.

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 Quiz
45

 Does a plane fly?

 Does a boat swim?
 Does a computer think?

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥

 About Myself
46

 H.D.R. “Accreditation to supervise PhDs” (in french ‘Habilitation à Diriger des Recherches’), Computer
Science, University Paris 13, (France), 1989-1997.
 PhD, Computer Science, University Paris 13, (France), 1986-1989.
 Master (DEA : Post-Graduate Diploma, 1 year), Computer Science, University Paris 6 (France), 1986.

Engineer (5 years), Computer Science, University of Algiers (Algeria), 1979-1984.
 September 2007-October 2009, full professor at the University of Artois (France).
 June 2007-October 2009, Researcher at the Laboratory of Computer Science of the
 University of Artois (CRIL, France).
 September 1992- June 2007, associate professor, first class, computer science, at the University of Cergy-
Pontoise (Paris).
 September 1992- June 2007, researcher at the Laboratory of Computer Science of the
 university Paris 13 (LIPN, Paris).
 1991-1992, associate professor, computer science at the university Paris 13.
 1990, consultant at INRIA (National Research Institute in Computer Science and Automatism) for the project
CAPRAN (in French ‘Calcul Parallèle Numérique'), Paris.

Artificial Intelligence : Introduction ١٤٤٣/٠٣/١٥