CS 4813 Artificial Intelligence

Shahzad Rafiq
Dept. of Computer Science,
MAJU, Islamabad

03/20/10 1
 Course Objectives
• Introduction to the basic principles, techniques,
and applications of Artificial Intelligence.
• Knowledge representation,
– logic, inferencing, problem solving, search algorithms,
game theory,
– perception, learning, planning, and agent design.

• Futuristic
– expert systems, neural networks, fuzzy logic, robotics,
natural language processing, and computer vision.

03/20/10 2
 Major Contents
• Logic
– Propositional Calculus, Predicate Calculus
• Neural Networks
– Perceptrons, Back Propagation, Self Organization
• Fuzzy Logic
• Evolutionary Algorithms
– Genetic Algorithm
• Agent Based Approaches

03/20/10 3
 Reference Book
• Textbook: Russell S. and Norvig P.,
Artificial Intelligence A Modern Approach,
2nd Ed, Prentice Hall 2003.

03/20/10 4
 Assessment Criteria
• Quizzes 10%
• Assignments & Project 30%
• Presentation 10%
• Mid Terms 30%
• End Term 20%

03/20/10 5
 Artificial Intelligence?

Lec-1

03/20/10 6
 What is Artificial Intelligence?
• What is AI?
• What is intelligence?
• What features/abilities do humans (animals?
animate objects?) have that you think are
indicative or characteristic of intelligence?

• abstract concepts, mathematics, language,

problem solving, memory, logical reasoning,
emotions, morality, ability to learn/adapt, etc…
03/20/10 7
Alternate Definitions (Russell + Norvig)
Like humans Not necessarily like humans

Systems that think like Systems that think rationally

humans
Think

Systems that act like Systems that act rationally

Act

humans

03/20/10 8
 Human intelligence
• Shall AI imitate (copy) humans or not?
• Advantage
– Humans are easy acknowledged exemplar of
intelligence.
• The Turing Test:
– A human interrogator: Communicates with a hidden
subject that is either a computer system or a human.
If the human interrogator cannot reliably decide
whether on not the subject is a computer, the
computer is said to have passed the Turing test.

03/20/10 9
 An application of the Turing Test

See Luis von Ahn, Manuel Blum, Nicholas Hopper, and John
Langford. CAPTCHA: Using Hard AI Problems for Security. In
Eurocrypt.

03/20/10 10
 Human intelligence
• Question: Is any System passing a turning
test is intelligent?

03/20/10 11
 Computer vs Brain (2003)
Computer Human Brain
Computational Units 1 CPU, 108 gates 1011 neurons

Storage Units 1010 bits RAM 1011 neurons

1011 bits disk 1014 synapses

Cycle time 10-9 sec 10-3 sec

Bandwidth 1010 bits/ sec 1014 bits/ sec

Memory updates/ sec 109 1014

In general there are various reasons why trying to mimic humans might
03/20/10 12
not be the best approach to AI.
 Imitate Human Intelligence or not?
• Lack of Knowledge - brain performance of
higher level processes
• Little information- for scientific
understanding of these processes
– Neuroscience has been very influential in
some areas of AI e.g.
• robotic sensing, vision processing, etc.

03/20/10 13
 Rationality
• Alternative approach

• Precise mathematical notion- do the right thing

in any particular circumstance

– A precise mechanism for analyzing and

understanding the properties of this ideal behavior we
are trying to achieve.

– A precise benchmark against which we can measure

the behavior the systems we build.
03/20/10 14
 Rationality
• Origin of Mathematical characterizations of rationality
such as
– logic (laws of thought) and
– economics (utility theory how best to act under uncertainty,
– game theory how self-interested agents interact

• But “No Silver Bullet“ solution exists

• We can study them and give exact characterizations of

their properties, good and bad.

03/20/10 15
 Computational Intelligence
• AI tries to understand and model intelligence as
a computational process.

• Thus we try to construct systems whose

computation achieves or approximates the
desired notion of rationality.

• Hence AI is part of Computer Science.

– Other areas interested in the study of intelligence e.g.,
– cognitive science: focuses on human intelligence.
related, but their central focus tends to be different

03/20/10 16
Alternate Definitions (Russell + Norvig)

Like humans Not necessarily like humans

Systems that think like Systems that think rationally
human
Think

Systems that act like Systems that act rationally

humans
Act

03/20/10 17
 Agency
• Focus on acting rationally
– which has implications for thinking/reasoning
• Useful to think of intelligent systems as being agents,
either:
– with their own goals
– or that act on behalf of someone (a “user”)

• An agent is an entity that exists in an environment and

that acts on that environment based on its perceptions of
the environment

• An intelligent agent acts to further its own interests (or

those of a user).
03/20/10 18
 Agent

Agent
perceives acts

Environment

• This diagram oversimplifies the internal

structure of the agent.
03/20/10 19
 Agent
prior knowledge user

Knowledge Agent Goals

acts
perceives
Environment
• Require more flexible interaction with the environment, the
ability to modify one’s goals, knowledge that be applied
flexibly to different situations.
03/20/10 20
 Degrees of Intelligence
• Dream: Building an intelligent system as capable
as humans (Humanoids)

• Systems have been built which exhibit various

degrees of intelligence e.g.

• Development of useful formalisms and algorithms

for construction of “intelligent” systems.
– forms the foundation of our attempt to understand
intelligence as a computational process.

03/20/10 21
 A U.S. Marine Corps technician
prepares to use a telerobot to detonate
ASIMO, a humanoid robot a buried improvised explosive device
manufactured by Honda near Camp Fallujah, Iraq

03/20/10 22
 Artificial Intelligence

03/20/10 23
 Applied Areas
Robotics

NLP

Search,
Expert Reasoning,
Systems Learning
Planning

Computer
Vision
o Disciplines
03/20/10 which form the core of AI- inner circle 24
o Fields which draw from these disciplines- outer circle.
 Applied Areas
• Game Playing
– Deep Blue Chess program
beat world champion Gary
Kasparov
• Speech Recognition
– PEGASUS spoken language
interface to American Airlines'
EASY SABRE reservation
system, which allows users to
obtain flight information and
make reservations over the
telephone.
03/20/10 25
 Applied Areas
• Mathematical Theorem Proving
– Use of inference methods to prove new theorems.
• Natural Language Understanding
– AltaVista automated translation of web pages.
– Translation of Catepillar Truck manuals into 20
languages.
• (Note: One early system translated the English sentence "The
spirit is willing but the flesh is weak" into the Russian equivalent
of "The vodka is good but the meat is rotten.")

03/20/10 26
 Computer Vision
• Face recognition programs in use by banks,
government, etc.
• The ALVINN system autonomously drove a van
from Washington, D.C. to San Diego (all but 52 of
2,849 miles), averaging 63 mph day and night,
and in all weather conditions.
• Handwriting recognition, electronics and
manufacturing inspection, photo-interpretation,
baggage inspection, reverse engineering to
automatically construct a 3D geometric model.

03/20/10 27
 Expert Systems
• Application-specific systems that rely on
obtaining the knowledge of human experts in an
area and programming that knowledge into a
system.

• DENDRAL, mass spectrometer interpreter

• MYCIN, modeling medical expert

• Microsoft Office Assistant:- customized help to

individual user
03/20/10 28
 Financial Decision Making
• Credit card companies, mortgage companies and
banks AI systems detect fraud
• expedite financial transactions like credit checks.
• profiles of customer usage patterns
• use profiles to detect unusual patterns take
appropriate action.

03/20/10 29
 Scheduling and Planning
• Automatic scheduling for manufacturing.
• American Airlines rerouting contingency
planner.
• European space agency planning and
scheduling of spacecraft assembly,
integration and verification

03/20/10 30
 Intelligent Robotics
• Robot Toys
– Aibo, I-Cybie, LEGO etc
• Robot Security
– Cye robot
• Robot Home Help
– Lawn mower, Vacuum cleaner etc

03/20/10 31
 Future: Autonomic Computing:
Concept CHOP

The Vision of Autonomic Computing, Jeffrey O.Kephart, David M.Chess, IBM Thomas J.Watson
03/20/10 32
Research, Retrieved on December 08, 2010,
http://www.research.ibm.com/autonomic/research/papers/AC_Vision_Computer_Jan_2003.pdf
 AI Challenges
• Translating telephone
• Accident-avoiding car
• Home help robot
• Smart clothes
• Intelligent agents that monitor and manage
information by filtering, digesting, abstracting
• Tutors
• Self-organizing systems, e.g., that learn to
assemble something by observing a human do
it.
03/20/10 33
 Summary
• Objectives of AI:
– Studies the design of systems that
{think, act} like {humans, rationally}.
• AI successful applications:
– decision making, expert systems, natural language
understanding, image processing.

• See Chapter 1 of Russell & Norvig.

03/20/10 34