Lecture 1 1

Habeebah A. Kakudi Ph.D.

11/25/21
qIntroduction to AI (Lecture 1)
ü Definition
ü History
ü Applications
üThe state of art and research areas
qIntelligent Agents (Lecture 2,3)
üAgents and environment
üGood behavior
ü The concept of rationality
ü The nature of environments
ü Structure of agents

11/25/21 2
qProblem Solving (Solving Problems by Searching
strategies) (lecture 4 ,5)
üUninformed Search
üBreadth-First Search(BFS)
ü Uniform-Cost Search(UCS)
ü Depth-First Search(DFS)
ü Depth-Limited Search(DLS)
üIterative deepening search
üHeuristic Search
ü Best First Search
ü Constraint Satisfaction Problems (CSP)
ü Game Tree Search
11/25/21 3
qRepresentation of knowledge (lecture 6)
üPropositional logic
üfirst order predicate logic
qKnowledge inference (lecture 7,8)
üProduction based system
ü Frame based system.
ü Inference - Backward chaining
ü Forward chaining
ü Rule value approach
üFuzzy reasoning
11/25/21 4
qExpert systems (lecture 9)
üArchitecture of expert systems,
ü Roles of expert systems
üKnowledge Acquisition ,Meta knowledge, Heuristics
ü Typical expert systems - MYCIN, DART, XOON
ü Expert systems shells.
qMachine Learning (Lecture 10)
üA General Model of Learning
ü Types of Learning Systems
ü Knowledge-Free Inductive Learning Systems
ü Learning from Single Examples

11/25/21 5
qFuzzy Logic (Lecture 11,12)
üWhat is Fuzzy Logic?
üFuzzy Logic Systems Architecture
üFuzzy Set Theory
üCrisp and Non-Crisp Set
ü Membership Function
üApplication Areas of Fuzzy Logic
qNeural Networks (Lectures 13 ,14)
üIntroduction, or how the brain works ?
ü The neuron as a simple computing element
ü The perceptron
üMultilayer neural networks
This is a very rough syllabus. It is almost certainly the case that
we will deviate from this. Some chapters will be treated only partially. 11/25/21 6
§Test at the 8th week 25%
§Course Works (Laboratory Assignment),
Attendance 5%
§Final Exam 70%

11/25/21 7
 Text
Text Book1:
Artificial Book:Artificial
Intelligence: A
Intelligence: A
Modern
Modern
Approach,
Approach,
Stuart J. Russell
Stuart J. Russell
and Peter Norvig
and Peter Norvig
, 3rd Edition
, 3rd Edition
2010
2010

AI - (2017-2018) -Diaa Eldein Mustafa - Lecture (0) -Introduction to Artificial

Intelligence 8
q I expect you to attend all the lectures. You must have 75% attendance (refer to your
GEAR handbook).
q You are expected to register on Moodle using the University domain email address.
q The lecture slide (PDF) presentations will be dropped in Moodle cover all the topics.
q These lecture slides do not contain much details of discussion.
q The lectures will consist of pdf slides, spoken material, and additional examples given
on the class.
q In order to understand the subject and the reasons for studying the material you will
need to attend the lectures and take notes to supplement lecture slides.
q This is your responsibility. If there is anything you do not understand during the
lectures, then ask, either during or after the lecture.
q In addition ,you must use the text book to supplement the lecture material by reading
around the subjects.

11/25/21 9
qIntroduction to AI
(Lecture 1)
ü Definition
ü History
ü Applications
üThe state of art and
research areas
11/25/21 10
qPhilosophers have been trying for over two
thousand years to understand and resolve
two big questions of the universe:
ü How does a human mind work ? And
ü Can non-humans (machines/ Software) have
minds?
q However, these questions are still unanswered

11/25/21 11
qSome philosophers have picked up the
computational approach originated by computer
scientists and accepted the idea that machines can
do everything that humans can do.
qOthers have openly opposed this idea, claiming that
such highly sophisticated behaviors such as love,
creative discovery and moral choice will always be
beyond the scope of any machine.

11/25/21 12
qThe nature of philosophy allows for disagreements to remain
unresolved.
qIn fact, engineers and scientists have already built machines
that we can call ‘intelligent’. So what does the word
‘intelligence’ mean?
qLet us look at a dictionary definition.
1 - Someone’s intelligence is their ability to understand and
learn things.
2 - Intelligence is the ability to think and understand instead
of doing things by instinct or automatically

11/25/21 13
q It is composed of :
üThinking (abstractly, using analogies)
üLearning and understanding from experience
üKnowledge applying successfully in new situations
üActing in complex environments,
ü Perceiving one’s environment
ü Automated behaviors.
ü Solve problems and to make decisions.
ü Pattern matching and recognition.
ü Reasoning (to solve problems and discover hidden knowledge)
ü Inference and prediction.
ü Perception.
ü Analysis .
ü Linguistic Intelligence.
üCreativity , Ingenuity , Expressive-ness, Curiosity.
11/25/21 14
11/25/21 15
“The exciting new effort to make computers think … machines with
minds,
in the full literal sense.”
Haugeland, 1985

“The study of mental faculties through the use of computational

models.”
Charniak and McDermott, 1985
“A field of study that seeks to explain and emulate intelligent behavior
in terms of computational processes.”
Schalkoff, 1990
“The study of how to make computers do things at which, at the moment,
people are better.”
Rich & Knight, 1991
11/25/21 16
§ Textbooks often define artificial intelligence as “the study and design of computing
systems that perceives its environment and takes actions like human beings”.
§ The term was introduced by John McCarthy in 1956 in the well-known Dartmouth
Conference.
§ In my study, an AI is defined as a system that possesses at least one (not
necessarily all) of the abilities mentioned in the previous page.

11/25/21 17
qKnowledge Representation (how does a program represent its
domain of discourse?)
qAutomated reasoning.
qPlanning (get the robot to find the bananas in the other room).
qMachine Learning (adapt to new circumstances).
qNatural language understanding.
qMachine vision, speech recognition, finding data on the web,
robotics, and much more.
 WHY STUDY AI?

Medicine/
Internet
Diagnosis
Search engines
Science
Labor
Science
Industry
Automation
Appliances Self-Driving What
else? 19
§ Representatives
§ George Boole
§ Alfred North Whitehead
§ Bertrand A. W. Russell

• PM was an attempt to describe a set of

axioms and inference rules in symbolic
§ Main contributions logic from which all mathematical truths
could in principle be proven
§ Boolean algebra
• However, in 1931, Gödel‘s incompleteness
§ Principia Mathematica theorem proved definitively that PM
could never achieve this lofty goal.

(from wikepedia)

11/25/21 20
§ Representatives
§ Alan Turing
§ Claude Shannon
§ John von Neumann

§ Main contributions
§ Theory of computation, Turing Machine
§ Turing test (to distinguish machine from human)
§ Information theory, application of Boolean algebra
§ von Neumann model of computing machines

11/25/21 21
§ Representatives
§ John McCarthy
§ Marvin Lee Minsky
§ Herbert Alexander Simon The term AI was proposed by these
§ Allen Newell
• persons in the well-known Dartmouth
§ Edward Albert Feigenbaum Artificial Intelligence conference
(1956)
§ Main contributions
§ LISP
§ Semantic network and frame
§ General problem solver and Expert systems

11/25/21 22
§ Representatives
§ David Rumelhart
§ Lotfi Zdeh
§ John Holland
§ Lawrence Forgel
§ Ingo Rechenber
§ John Koza
§ Main contributions
§ Learning of MLP
§ Fuzzy logic
§ Genetic algorithms
§ Evolutionary programming
§ Evolution strategy
§ Genetic programming

Soft computing
Human like computing and natural computing
11/25/21 23
§ Representative technologies
§ Internet
§ Tim Berners-Lee, WWW inventor, 1989 –
§ Internet of things
§ Kevin Ashton, MIT Auto-ID Center, 1999
§ Cloud computing
§ Main frame (1950s), virtual machine (1970s), cloud (1990s)
§ Big data
§ John R. Masey, SGI, 1998
§ Deep learning
§ Geoffrey Hinton, UoT, 2006

11/25/21 24
§ In March 2016, Alpha-Go of DeepMind defeated Lee Sedol, who was the strongest
human GO player at that time.
§ This is a big news that may have profound meaning in the human history.

11/25/21 25
• We will follow “act rationally”
approach
– Distinction may not be that
important
• acting rationally/like a human
presumably requires (some sort of)
thinking rationally/like a human,
• humans much more rational
anyway in complex domains

11/25/21 26
qBuild intelligent artifacts vs. understanding human
behavior.
qDoes it matter how I built it as long as it does the
job well?
qShould the system behave like a human or behave
intelligently?

The Turing
Test
§ A serious science.
§ General-purpose AI like the robots of science fiction is incredibly
hard
§ Human brain appears to have lots of special and general functions, integrated
in some amazing way that we really do not understand at all (yet)
§ Special-purpose AI is more doable (nontrivial)
§ E.g., chess/poker playing programs, logistics planning, automated translation,
voice recognition, web search, data mining, medical diagnosis, keeping a car
on the road, … … … …

11/25/21 28
§ 1950 – Alan Turing devised a test
for intelligence called the Imitation
Game
§ Ask questions of two entities, receive
answers from both
§ If you can’t tell which of the entities is
human and which is a computer Questions
program, then you are fooled and we
should therefore consider the computer Answers
Answers
to be intelligent
§ Loebner prize
§ Current programs nowhere close to
passing this
§ http://www.jabberwacky.com/
§ http://turingtrade.org/ Which is the person?
Which is the computer?
§ Created by Joseph Weizenbaum
§ This system would act like a Rogerian psychoanalyst by
holding a dialog with a person
§ the dialogwould be consist of the doctor (Eliza) asking
questions, the human responding, and the doctor using the
response to ask another question
§ Weizenbaum wrote the program so that it would
generate an English response/question based on a
group of patterns
§ If the user sentence matched a pattern, this pattern would be
used to generate the next sentence/question. See ELIZA
Chatterbot
§ A more recent, and much superior version is called A.L.I.C.E,
see Alice Chatterbot
§ click on “chat with A.L.I.C.E.”
– H: Men are all alike
– E: In what way?
– H: They’re always bugging us about something or another.
– E: Can you be more specific?
– H: Well, my boyfriend made me come here.
– E: Your boyfriend made you come here?
– H: He says I’m depressed much of the time.
– E: I am sorry to hear you are depressed.
– H: It’s true. I am unhappy.
– E: Do you think coming here will help you not to be unhappy?
– H: I need some help, that much seems certain.
– E: What would it mean to you if you got some help?
– H: Perhaps I could learn to get along with my mother.
– E: Tell me more about your family
Jerry: my name is jerry.
ALICE: Hey jerry.
Jerry: do you know Turing?
ALICE: A brilliant and tragic figure in the history of computer science.
Jerry: who else do you know?
ALICE: Hugh Loebner, Robby Garner, and Steven Spielberg
Jerry: which film do you like?
ALICE: My favorite movie is Teknolust. Have you seen it?
Jerry: no.
ALICE: I would recommend you check it out.
Jerry: what is it about?
ALICE: Science fiction.
Jerry: do you remember my name?
ALICE: Your name is jerry, seeker.

Notice unlike Eliza, ALICE has a memory, can handle references

(have you seen it? no. I would recommend it. See https://en.wikipedia.org/wiki/Artif
§ With Eliza or Alice like rules, we can eventually solve
the Turing Test – it just takes writing enough rules
§ Does the system understand what it is responding to?
§ No, neither Eliza nor Alice understand the text, its just that
Alice has better, more in depth and wider ranging rules
§ However, we could build a representation that models some real-world
domain and knowledge base
§ The system can fill in information from the conversation
§ this is sort of like a database, or an object with data attributes to be
filled in
§ we can use a variety of AI representations like scripts, frames,
semantic networks
§ Questions can be responded to by looking up the stored data
§ In this way, the system is responding, not based merely on
“canned” knowledge, but on knowledge that it has “learned”
§ Does this imply that the system knows what it is
discussing?
§ What does it mean to know something?
“
§ By John Searle
§ Person who knows English but not Chinese
image from http://www.unc.edu/~prinz/pictures/c-room.gif

sits in room
§ Receives notes in Chinese
§ Hassystematic English rule book for how to
write new Chinese characters based on input
Chinese characters, returns his notes
§ Person=CPU, rule book=AI program, really also need lots of
paper (storage)
§ Has no understanding of what they mean
§ But from the outside, the room gives perfectly
reasonable answers in Chinese!

§ Searle’s argument: the room has no intelligence in

it!
11/25/21 34
qThe ability of a system to :
ü Calculate
ü Reason
ü Perceive relationships and analogies
ü Learn from experience
ü Store and retrieve information from memory
ü Solve problems, comprehend complex ideas
ü Use natural language fluently
ü Classify, generalize, and adapt new situations.

11/25/21 35
 Natural lang.
input effectors
vision

Knowledge learning
representation

planning reasoning
§ Early work
§ Theoretic foundations

§ First wave:
§ Reasoning with given knowledge

§ Second wave
§ Learning-based knowledge acquisition

§ Third wave
§ Learn in the cyber-space

11/25/21 37
11/25/21 38
§ Clearly-defined tasks
that we think require intelligence and
education from humans tend to be doable for AI techniques
§ Playing chess, drawing logical inferences from clearly-stated facts, performing
probability calculations in well-defined environments, …
§ Although, scalability can be a significant issue

§ Complex, messy, ambiguous tasks that come natural to humans (in

some cases other animals) are much harder
§ Recognizing your grandmother in a crowd, drawing the right conclusion from an
ungrammatical or ambiguous sentence, driving around the city, …

§ Humans better at coming up with reasonably good

solutions in complex environments
§ Humans better at adapting/self-evaluation/creativity
(“My usual strategy for chess is getting me into trouble
against this person… Why? What else can I do?”)
11/25/21 39
§ Search and optimization
§ Knowledge representation
§ Reasoning and automatic proving
§ Learning and understanding
§ Pattern classification / recognition
§ Planning
§ Problem solving

11/25/21 40
§ Brute-force search
§ Depth-first search
§ Breadth-first search
§ Heuristic Search
§ Hill climbing search
§ Best-first search
§ A* Algorithm
§ Intelligent search
§ Genetic algorithms
§ Meta-heuristics

11/25/21 41
§ What is the input?
§ Map from real world to the mind model

§ What is the output?

§ Map from the mind model to the real world

§ What is the relation between the input and the output?

§ Abstraction of the real world

11/25/21 42
§ Representation of the problem
§ State space representation
§ Vector representation
§ Representation of knowledge
§ Production (decision) rules
§ Semantic network and ontology
§ Predicate logic
§ Fuzzy logic
§ Neural network (for tacit knowledge)

11/25/21 43
§ Neural network learning
§ Including MLP, SVM, deep learning, etc.
§ Evolutionary learning
§ GA or meta-heuristics in general
§ Reinforcement learning
§ Artificial immune system
§ Fuzzylogic
§ Decision tree
§ Hybrid system
11/25/21 44
§ Learn how to use the basic search methods;

§ Understand the basic methods for problem formulation and knowledge

representation;
§ Understand the basic idea of automatic reasoning;
§ Know some basic concepts related to pattern recognition and machine learning.

11/25/21 45