Course Outline
The course consists of:
COMP210: Artificial Intelligence
30 lectures slots (may use some for tutorials);
References
(outlined in the course guide)
Good AI books include:S. Russell and P. Norvig. AI A Modern Approach.
Second Edition Prentice Hall, 2003
tutorial exercises;
Lecture 1. Introduction
lab exercises;
Not assessed
Class test based on the practicals!!
Boris Konev
http://www.csc.liv.ac.uk/∼konev/COPM210/
M. Ginsberg. Essentials of Artificial Intelligence.
Morgan Kaufmann, 1993.
E. Rich and K. Knight. Artificial Intelligence,
McGraw-Hill, 1991 (2nd edition)
enough self study to understand the material;
two assignments;
The following is a (cheap) recent text (not as good as the
above) covers standard material.
a two hour exam.
A. Cawsey. The Essence of Artificial Intelligence.
Prentice-Hall, 1998.
Course materials, syllabus, the course guide, lecture slides,
tutorial and lab exercises etc can be obtained from
http://www.csc.liv.ac.uk/∼konev/COMP210
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.1/21
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.2/21
References (contd.)
Boris Konev
Learning Outcomes
Course Contents
The following is a Prolog book.
Introduction to Artificial Intelligence
An awareness of the principles of knowledge
representation.
Prolog - an AI programming language
I. Bratko. Prolog Programming for Artificial Intelligence.
Addison Wesley 1990.
Search
An understanding of search techniques and logic,
particularly as related to knowledge representation.
Knowledge Representation
Propositional Logic
An understanding of the major knowledge representation
paradigms: production rules, prepositional and first order
predicate calculus and structured objects.
First-Order Logic
Resolution Based Proof for Propositional and
First-Order Logics
An understanding of how these representations can be
manipulated to solve problems in a knowledge based
systems context.
Expert Systems
AI Applications
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.4/21
Boris Konev
Learning Outcomes (contd.)
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.3/21
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.5/21
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.6/21
What I expect from you.
Credits
Some appreciation of the major knowledge based
systems.
To attend lectures.
Awareness of other applications of AI.
To turn mobile phones off and not to chat in lectures.
Familiarity with the essentials of Prolog so as to enable
exploration of the above in practice.
To do whatever reading and self study is required to
understand the material.
Clare Dixon
To attempt the tutorial and laboratory exercises.
Michael Wooldridge
To be punctual.
This set of slides is based on the materials provided by
people who used to teach this course in the University of
Liverpool
Simon Parsons
To carry out assessed work individually and hand it in
on time.
Handing in assessed work is very important.
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.7/21
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.8/21
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.9/21
What is intelligence?
What is AI?
For thousands of years people tried to understand
how we think
Philosophy
Four Views of AI
AI attempts to build intelligent entities
1. AI as acting humanly — as typified by the Turing test
AI is both science and engineering:
2. AI as thinking humanly — cognitive science.
3. AI as thinking rationally — as typified by logical
approaches.
the science of understanding intelligent entities — of
developing theories which attempt to explain and
predict the nature of such entities;
the engineering of intelligent entities.
Mathematics
What is correct mathematical reasoning?
Neuroscience
4. AI as acting rationally — the intelligent agent approach.
Psychology
Economics
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.10/21
Boris Konev
Acting Humanly
Most famous response due to Alan Turing, British
mathematician and computing pioneer:
Expert systems — “AI success story in early 80’s”
No program has yet passed Turing test!
(Annual Loebner competition & prize.)
Try to understand how the mind works — how do we
think?
A program that succeeded would need to be capable of:
Two possible routes to find answers:
by introspection — we figure it out ourselves!
by experiment — draw upon techniques of
psychology to conduct controlled experiments. (“Rat
in a box”!)
The discipline of cognitive science: particularly
influential in vision, natural language processing, and
learning.
Note no visual or aural component to basic Turing test
— augmented test involves video & audio feed to entity.
Boris Konev
Human vs Machine Thinking (I)
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.14/21
Boris Konev
Human vs Machine Thinking (II)
Trying to understand how we actually think is one route
to AI — but how about how we should think.
“Human way”
Tried by World champion M.Botvinnik (who also was
a programmer)
Rule-based representation of knowledge
Typical domains are:
medicine (INTERNIST, MYCIN, . . . )
geology (PROSPECTOR)
chemical analysis (DENDRAL)
configuration of computers (R1)
Use logic to capture the laws of rational thought as
symbols.
Reasoning involves shifting symbols according to
well-defined rules (like algebra).
Poor performance
Result is idealised reasoning.
“Computer way”
Sophisticated search algorithms
Vast databases
Immense computing power
Thinking humanly works
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.15/21
Thinking Rationally
Computer program playing chess
Human expert’s knowledge and experience is passed to
a computer program
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.12/21
Thinking Humanly
natural language understanding & generation;
knowledge representation;
learning;
automated reasoning.
Human interrogates entity via teletype for 5
minutes. If, after 5 minutes, human cannot tell
whether entity is human or machine, then the
entity must be counted as intelligent.
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.13/21
Boris Konev
Turing test
Emphasis on how to tell that a machine is intelligent,
not on how to make it intelligent
when can we count a machine as being intelligent?
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.11/21
Human world champion beaten
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.16/21
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.17/21
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.18/21
Logic and AI
Acting Rationally (I)
Logicist approach theoretically attractive.
Acting Rationally (II)
Acting rationally = acting to achieve one’s goals, given
one’s beliefs.
Lots of problems:
Achieving perfect rationality — making the best
decision theoretically possible — is not usually possible,
due to limited resources:
An agent is a system that perceives and acts; intelligent
agent is one that acts rationally w.r.t. the goals we
delegate to it.
transduction — how to map the environment to
symbolic representation;
representation — how to represent real world
phenomena (time, space, . . . ) symbolically;
reasoning — how to do symbolic manipulation
tractably — so it can be done by real computers!
limited time;
limited computational power;
limited memory;
limited or uncertain information about environment.
Emphasis shifts from designing theoretically best
decision making procedure to best decision making
procedure possible in circumstances.
The trick is to do the best with what you’ve got!
Logic may be used in the service of finding the best
action — not an end in itself.
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.19/21
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.20/21
Boris Konev
COMP210: Artificial Intelligence. Lecture 1. Introduction – p.21/21