Artificial Intelligence

An Introduction to
Artificial Intelligence
Dr Iman Ardekani

Understanding intelligence
Imitating intelligence
Artificial intelligence - AI
Branches of AI
Content

Undrstanding Intelligence
Many great philosopher over the ages attempted to explain the
process of thought and understanding.
Intelligence
Plato
428–348 BC
Aristotle
384–322 BC
Philosophy
Math
Nature and Universe
Human knowledge Intelligence
*

Plato
428–348 BC
Aristotle
384–322 BC
Copernic
1473-1543
Galileo
1564-1642
Philosophy & Natural Science
Math
Intelligence
Nature and Universe
Human knowledge
The real key that started the quest for the
simulation of inteligence did not occure until …
*

Philosophy & Natural Science
Math
Intelligence
Nature and Universe
Human knowledge
Thomas Hobbes (English Philosopher) put forth an interesting concept that
thinking consists of symbolic operations and that everything in the life
can be represented mathematically.
Hobbes
1588-1679
*

Hobbes (British Philosopher):
Thinking consists of symbolic operations!
Based on this logic, a machine capable of caring out
mathematical operations on symbols could imitate human
thinking.
Hobbes
1588-1679
What is a symbolic operation?
• Numeric operation (2+3)2 = 25
• Symbolic operation (a+b)2 = a2 + b2 + 2ab

Rene Descartes (French Philosopher and Mathematician):
He believed that the mind and the real world are in parallel
planes. The physical word (i.e. machines) cannot imitate the
mind because there is no common reference point.
Descartes
1596-1650

Charles Babbage (British Mathematician):
In Babbage's time, numerical tables were calculated by humans who
were called 'computers’. He saw the high error-rate of this human-
driven process and started work of trying to calculate the tables
mechanically. He created a “difference engine” to compute values
of polynomial functions.
Imitating Inteligence
Babbage
1791-1871
A part of Babbage's difference
engine
He also introduced the idea of “Analytical
Machine”, but he could never realize this
idea.

George Boole (British Mathematician):
Boole formulated the “Laws of Thought” that set up rules of logic for
representing thoughts (symbolic logic). This was the birth of digital
logic, a key component of AI.
In the early 1900s, Alfred Whitehead and Bertrand Russell extended
Boole’s logic to include mathematical operations. This led to the
formulation of digital computers. Also, this made possible one of the
first ties between computers and thought process.
Boole
1815-1864
Russell
1872-1970
Whitehead
1861-1947

Design a digital computer using logical operations to compute y=x1+x2
where x1 and x2 are 4-digit binary numbers (4-bit adder).
Design a digital computer using logical operations to compute y=x1.x2
where x1 and x2 are 4-digit binary numbers (4-bit multiplier).
Design a digital computer using logical operations to compute y=ex where
x1 and x2 are 4-digit binary numbers (ex=1+x+x2/2+x3/6+…).

Claude Shannon (American Electrical Engineer):
He wrote his master’s thesis demonstrating that electrical
applications of Boolean algebra could construct and resolve any
logical, numerical relationship.
It has been claimed that this was the most important master's thesis
of all time. His PhD these was on mathematical relationships of
genetics.
He is known as the father of Information Technology.
Shannon
1916-2001

John Neumann (American Mathematician)
He suggested that the computers
 should be general purpose logic machines.
 could react intelligently to the results of their calculations
 could choose among alternatives, and even play checker and chess
This represented something unheard of at that time: a machine with
built-in intelligence, able to operate on internal instructions.
Before introducing this concept, even the most complex mechanical
devices had always been controlled from the outsides, by knobs and
dials.
He didn't’ invent the computer but what he introduced was equally
significant: computing by use of computer programs.
Neumann
1903-1957

John Mauchly (American Electrical Engineer):
John Mauchly designed and built the first general purpose digital
computer in 1946 at the University of Pennsylvania:
ENIAC (Electronic Numerical Integrator and Computer)
Weight = 30 Tons
Floor Space = 1500 Square Feet
Shannon’s idea  Hardware
Neumann’s idea  Software
Mauchly
1907-1980

Alan Turing (British Mathematician):
He introduced “Universal Machine Concept” that describe a
machine for solving all problems based on variable instructions.
Turing’s universal machine concept, along with Neumann’s concept
of computing using programs led to programmable computers.
Operational machines were now being realized. The question was
“Are they intelligent?” and “in what extend?”. Turing also designed
Turing’s test for determining the intelligence of a system.
Turing
1912-1954

Turing Test – Step 1 (man/woman)
A is a man and B is a woman and C is of either sex.
C is unable to see either A or B, and can communicate
with them only through online computer chat.
By asking questions of A and B, C tries to determine
which of the two is the man and which is the woman.
A's role is to trick C into making the wrong decision,
while B attempts to assist C in making the right one.

Turing Test – Step 2 (human/computer)
Substitute a computer for A.
By asking questions of Computer and B, C tries to
determine which of the two is the computer.
Computer's role is to trick C into making the wrong
decision, while B attempts to assist C in making the right
one.
If the C’s success rate in human/computer game is not
better than his success rate in the man/woman game

Turing Test
If the C’s success rate in human/computer game is not better than his
success rate in the man/woman game, then the computer can be said to be
“thinking”.

There was now a need for a high-level programming language.
Logic Theorist was written in 1955 by A. Newell, H. A. Simon
and J. C. Shaw. It was the first program deliberately
engineered to mimic the problem solving skills of a human
being and is called "the first artificial intelligence program.” It
would eventually prove 38 of the first 52 theorems of
Whitehead and Russell, and find new and more elegant proofs
for some.[2]

John McCarthy (American Computer Scientist)
He coined the term “Artificial Intelligence” in the first conference on
machine intelligence, 1956.
He also developed LISP (List Processing) programming language,
which has become a standard tool for AI development.
LISP distinctions:
 Memory organization – in a tree fashion
 Control structure – instead of working from perquisites to a goal, it
starts with the goal and works backward to determine what perquisites
are required to achieve the goal.
McCarthy
1927-2011

GPS (General Problem Solver) was another AI programming language that
introduced in 1959.
It was capable of solving theorems, playing chess, or doing puzzles.
Its core was based on the use of means-end analysis, which involves
comparing a present state with a goal state. The difference between the two
state is determined and a search is done to find a method to reduce this
difference. This process is continued until there is no difference between the
current state and the goal state.
It was capable of backtracking to an earlier state to correct its mistakes.
It was also able to define sub-goals.
GPS did a good job of imitating the human subjects.

ELIZA was the first intelligent computer program that
was enable of interacting in a two-way conversation.
It could sustain very realistic conversations by very
smart techniques.
For example, ELIZA used a pattern matching method
that would scan for keywords like “I”, “You”, “Like”
and so on. If one of these words was found, it would
execute rules associated with it. If no match was
found, it would request for more information.
Link to ELIZA

The various attempts at formally defining the use of machines to simulate
human intelligence let to several AI branches
1. Natural Language Processing (NLP)
2. Computer Vision
3. Robotics
4. Problem-solving and planning
5. Learning
6. Expert Systems
Branches of AI

Branches of AI
NLP
ComputerVision
ExpertSystems
ProblemSolving
Robotics
Learning
Human-like artificial creatures
Other artificial creatures
Special robots/machines with higher capabilities

How successful we have been in creating human-like artificial creatures?
Branches of AI

Natural Language Processing (NLP)
NLP understands, and generates languages that humans use naturally so that
eventually you will be able to address your computer as though you were
addressing another person (e.g. ELIZA)
Branches of AI
Speech NLP Knowledge

Natural Language Processing (NLP)
NLP Categories:
1- Phonology: modeling the pronunciation of words (chair, car, cell)
2- Morphology: identifying the structure of words (dog, dogs, hot dogs)
3- Syntax (identifying grammars)
4- Semantics (understanding and representing the meaning)
Applications:
automatic text indexing, grammar and style analyser, automatic text generation,
machine translation, optical character recognition (OCR) and etc.
Branches of AI

Computer Vision
Computer vision is a field that includes methods for acquiring, processing,
analysing, and understanding images and, in general, high-dimensional
data from the real world in order to produce numerical or symbolic
information, e.g., in the forms of decisions.
Branches of AI
Images
Computer
Vision
Knowledge

Branches of AI
Computer Vision
US Deference Advance Research Projects Agency (DARPA)

Computer Vision
Applications:
1. Recognize objects (e.g. people we know and things we own)
2. Locate objects in space (to pick them up?)
3. Track objects in motion (catching a baseball, avoiding collisions
with cars on the road)
4. Recognize actions (e.g. walking, running, pushing)
Branches of AI

Robotics
Robotics involves the control of actuators on robots to move, manipulate or
grasp objects, locomotion of independent machines and use of sensory
input to guide actions.
Branches of AI

Problem-solving and Planning
This technology involves application such s refinement of high-level goals
into lower-level ones, determination of actions to achieve goals, revision of
plans based on intermediate results, and focused search of important
goals. A good example is chess players software.
Branches of AI

Learning
Learning deals with research into various forms of learning including rote learning,
learning through advise, learning by example, learning by task performance, and
learning by following concepts.
Branches of AI

Expert Systems
Expert systems deal with the processing of knowledge as opposed to
processing of data. It involves the development of computer software to
solve complex decision problems. In fact, an expert system is a computer
system that make decisions on behalf of human.
Branches of AI
Link to ANNA Android Doctor

Artificial Intelligence

Empfohlen

Empfohlen

Weitere ähnliche Inhalte

Was ist angesagt?

Was ist angesagt? (20)

Ähnlich wie Artificial Intelligence

Ähnlich wie Artificial Intelligence (20)

Mehr von Iman Ardekani

Mehr von Iman Ardekani (8)

Kürzlich hochgeladen

Kürzlich hochgeladen (20)

Artificial Intelligence

Hinweis der Redaktion