1. ARTIFICIAL INTELLIGENCE
Department Of Information Technology, University Of Haripur, Khyber Pakhtunkhwa, Pakistan
» 1 INTRODUCTION
DR. YOUSAF SAEED
yousaf@uoh.edu.pk
2. Assessment
▪ Artificial Intelligence: Structures and Strategies for Complex Problem Solving: International Edition
by George F. Luger, 6th Edition: Pearson Education, 2008
▪ Artificial Intelligence: A Modern Approach by Stuart Jonathan Russell, Peter Norving, John F. Canny, 2nd
Edition, Prentice Hall, 2003.
▪ A First Course in Artificial Intelligence by Deepak Khemani
▪ Library, conference papers, journal articles ..
✔ Attendance
✔ Quizzes
✔ Assignments
✔ Presentations
✔ Timing and Deadlines
✔ Class Involvement
✔ Recent Research work
✔ Programming
✔ Discussions
✔ Small Projects
Sessional 25%
Mid-Term 25%
Final Term 50%
4. What AI is doing Today ?
❑ Penetrating in every direction
▪ Robotic Vehicles - STANLEY, Google Car, Stanford’s Car, BOSS ..
▪ Speech Recognition – Recognition of voice
▪ Autonomous Planning and Scheduling – Planning and Scheduling Tasks
o NASA’s Remote Agent Program controls scheduling of operations for space crafts
▪ Game Playing – Strategy games, Interactive games ..
o IBM’s Deep Blue first computer program defeated World champion Garry Kasprov in a Chess match ..
▪ Spam Fighting - Learning algorithms as spammers continuously updating their tactics.
▪ Logistics Planning - Dynamic Analysis and Replacing Tool (DART) logistics planning and scheduling for
transportation .. 50,000 vehicles, cargo and people at a time .. Starting point, destination, routes, conflict resolution
by US during Gulf War ..
▪ Project Maven - Designed to Deploy Computer Algorithms to War Zone.
▪ Robotics - Robotic vacuum cleaners, prosthetic arms ..
▪ Machine Translations - Program to automatically translate between languages.
▪ Innovation DX - for bringing data analytics to medicine.
▪ Search Engines – Google ..
..
7. ▪ If we say the given program thinks like human .. it must be determined how humans think
▪ (Necessity to get inside the actual workings of human mind)
▪ THREE Ways:
✔ Introspection (catching our own thoughts as they go by)
✔ Psychological Experiments (observing a person in action)
✔ Brain Imaging (observing the brain in action)
▪ Once we have sufficiently precise theory of mind, it is possible to express the theory as a computer
program.
▪ If program’s I/O behavior matches human behavior, that is evidence that some of program’s mechanism
could also be operating in humans.
▪ Cognitive Science – Computer models from AI and experimental techniques from psychology to
construct precise and testable theory of mind.
Argument in the AI field that algorithm performs well on a certain task and, therefore, is a good model of
human performance. Modern authors separate the two kind of claims.
Both AI and Cognitive Science continue to fertalize each other
Thinking Humanly – Cognitive Modeling Approach
8. ▪ Turing Test – Proposed by Alan Turing in 1950 provided satisfactory operational def., of intelligence.
▪ Computer passes the test if human interrogator after poses some written questions cannot tell whether
the written responses came from a person or a computer.
▪ Intelligent Behavior defined by Turing:
‘ability to achieve human-level performance in all cognitive tasks, sufficient to fool an interrogator’
▪ Computer would need to pass Turing Test by having,
Natural Language Processing + Knowledge Representation + Automated Reasoning + Machine Learning
▪ Turing Test avoided direct Physical interaction between the interrogator and the computer
▪ As physical simulation of a person is unnecessary for intelligence.
▪ However:
▪ Total Turing Test – It include video signal so that interrogator can test the subject’s perceptual abilities
as well as to pass physical objects.
▪ Computer would need to pass Total Turing Test by having,
Computer Vision (to perceive objects) + Robotics (to manipulate objects and move them)
Acting Humanly – Turing Test Approach
9. Turing Test
▪ Alan Turing (1950s) prescribed a test to determine whether a machine is intelligent !
▪ The Imitation Game
▪ Man (A), Woman (B) and Interrogator (C) who communicate only via messages.
▪ The interrogator (C) knows A and B as X and Y.
▪ (C) task is to correctly label X and Y as A or B byasking them questions.
▪ If (C) is bluffed by (A) or (B) as machines, we could say the machine is intelligent.
▪ Turing 1950 Paper:
“ I believe that in about fifty years time, it will be possible to program computers, with a storage
capacity of about 109, to make them play the imitation game so well that an average interrogator
will not have more than 70% chance of making the right identification after five minutes of
questioning .. I believe that at the end of century, the use of words and general educated opinion will
have altered so much, that one will be able to speak of machines thinking without expecting to be
contradicted. ”
10. Turing Test ..
▪ ELIZA – chatterbot program written by Joseph Weizenbaum 1966.
▪ “Responses of non-directional psychotherapist in an initial psychiatric interview”
▪ https://www.masswerk.at/elizabot/
▪ Other sources:
▪ http://www.med-ai.com/models/eliza.html
▪ Turing 1950 Paper:
“ I believe that in about fifty years time, it will be possible to program computers, with a storage
capacity of about 109, to make them play the imitation game so well that an average interrogator
will not have more than 70% chance of making the right identification after five minutes of
questioning .. I believe that at the end of century, the use of words and general educated opinion will
have altered so much, that one will be able to speak of machines thinking without expecting to be
contradicted. ”
11. ▪ In laws of thought approach, the emphasis is on Correct Inferencing.
To draw conclusion that this action will achieve the goal and start acting on it.
▪ Greek philosopher, Aristotle, one of the first to attempt to codify ‘right thinking’.
▪ His syllogisms (a kind of logical argument) provided patterns for argument structures
Socrates is a man; all men are mortal; therefore, Socrates is mortal
▪ These laws of thought were supposed to govern the operation of mind, their study initiated the field
called logic.
▪ Logicians in 19th century developed a precise Notation for statements of all kinds of things in the
world and relations between them.
▪ Although if No Solution exists, the program might loop forever.
▪ By 1965, programs existed that could solve any solvable problem described in logical notation.
Logicist in AI hopes to build on such programs to create intelligent systems.
Thinking Rationally – The Laws of Thought Approach
12. ▪ Agent – Something that acts (comes from Latin agere i.e. to do)
▪ Actually all computer programs are meant to do something but computer agents are expected to do more
Operate anonymously + perceive their environment + persist over prolong period of
time +adapt to change + create and pursue goals ..
▪ Rational Agent – Acts to achieve the best outcome, and when there is uncertainty, the best expected
outcome.
▪ There are ways of Acting Rationally that does not involve Inferences;
Pulling one’s hand off of a hot stove is a reflex action that is more
successful than a slower action taken after careful deliberation.
▪ Knowledge Representation, Reasoning - enable agents to take rational decisions
▪ Advantages of Rationality:
+ More general than the laws of thought approach
+ More open for scientific development
Acting Rationally – The Rational Agent Approach
13. AI Foundations
▪ Computer Engineering (branch of engineering that integrates electronic
engineering and computer sciences)
o How can we build an efficient computer? ..
▪ Philosophy (study of the fundamental nature of knowledge, reality, and
existence)
o Where does knowledge come from? How does mind arise from the physical
brain?
o How does knowledge lead to action? Can formal rules be used to draw
conclusions? ..
▪ Mathematics (science of structure, order and relation)
o What are the formal rules to draw valid conclusions? What can be computed?
How do we reason with uncertain information? ..
14. AI Foundations
▪ Neuroscience (study of the nervous system particularly the brain)
o How do brain process information? ..
15. AI Foundations ..
▪ Psychology (scientific study of mind and behavior)
o How do humans and animals think and act? ..
▪ Cybernetics (science of control and communications in the animal and machine)
o How can artifacts operate under their own control? ..
▪ Linguistics (scientific study of language and its structures)
o How does language relate to thought? ..
▪ Economics (social science concerned with the production, distribution and consumption of goods and services)
o How should we make decisions so as to maximize payoff? How should we do this
when other may not go along? How should we do this when payoff may be far in the
future? ..
16. Historic Backdrop
▪ The urge to create intelligent beings is traced back to Greek Mythology.
▪ Plato – Both an object in the world and idea of that object were derived from a world of perfect ideas.
▪ Our own ideas and objects in the world, could be imperfect.
▪ Aristotle – Ideas were like pictures of the objects.
▪ Human mind – Mirror reflecting the world to us.
▪ “I” itself is the creation of mind.
▪ Copernicus – Daily movement of the sun, moon, stars is illusion created by the earth rotation.
▪ Galileo – What we think and perceive is something that happens within us.
o Created representations: right angle triangle; speed and time, area represents distance.
o All reality is mathematical and everything is made up of particles.
▪ Hobbes – Thinking is the manipulation of symbols.
▪ Descartes – Dualist, separated mind from the body & Mind manipulate mental symbols.
▪ Idealism: There are only ideas and matter is something we imagine.
▪ Materialism: Only the material world exists, mental world is just a construct arises from
physical activity.
▪ David Hume – Just as the physical world operated according to the laws of nature, the mental
world also operated according to its own laws.