Advanced Intelligent Systems

First Slide, Introduction

Thomas Anung Basuki, Jan 2014

AIS - Intro p. 1

Course Goals
to understand the concept of AI and its relationships with computing

AIS - Intro p. 2

Course Goals
to understand the concept of AI and its relationships with computing to formulate a solution to a computational problem by using search techniques

AIS - Intro p. 2

Course Goals
to understand the concept of AI and its relationships with computing to formulate a solution to a computational problem by using search techniques to use knowledge representation, reasoning and machine learning techniques in solving problems

AIS - Intro p. 2

Course Goals
to understand the concept of AI and its relationships with computing to formulate a solution to a computational problem by using search techniques to use knowledge representation, reasoning and machine learning techniques in solving problems to analyse AI solutions

AIS - Intro p. 2

Course Overview
Introduction: AI concepts and Intelligent Agents

AIS - Intro p. 3

Course Overview
Introduction: AI concepts and Intelligent Agents Solving Problems by Searching

AIS - Intro p. 3

Course Overview
Introduction: AI concepts and Intelligent Agents Solving Problems by Searching Blind Search, Informed Search and Local Search

AIS - Intro p. 3

Course Overview
Introduction: AI concepts and Intelligent Agents Solving Problems by Searching Blind Search, Informed Search and Local Search Multiple Agents Search

AIS - Intro p. 3

Course Overview
Introduction: AI concepts and Intelligent Agents Solving Problems by Searching Blind Search, Informed Search and Local Search Multiple Agents Search Adversarial Search and Game Playing

AIS - Intro p. 3

Course Overview
Introduction: AI concepts and Intelligent Agents Solving Problems by Searching Blind Search, Informed Search and Local Search Multiple Agents Search Adversarial Search and Game Playing Knowledge Representation, Reasoning and Planning

AIS - Intro p. 3

Course Overview
Introduction: AI concepts and Intelligent Agents Solving Problems by Searching Blind Search, Informed Search and Local Search Multiple Agents Search Adversarial Search and Game Playing Knowledge Representation, Reasoning and Planning Logic-based Intelligent Systems and Reasoning

AIS - Intro p. 3

Course Overview
Introduction: AI concepts and Intelligent Agents Solving Problems by Searching Blind Search, Informed Search and Local Search Multiple Agents Search Adversarial Search and Game Playing Knowledge Representation, Reasoning and Planning Logic-based Intelligent Systems and Reasoning Planning

AIS - Intro p. 3

Course Overview
Introduction: AI concepts and Intelligent Agents Solving Problems by Searching Blind Search, Informed Search and Local Search Multiple Agents Search Adversarial Search and Game Playing Knowledge Representation, Reasoning and Planning Logic-based Intelligent Systems and Reasoning Planning Other Knowledge Representation Languages

AIS - Intro p. 3

Course Overview
Introduction: AI concepts and Intelligent Agents Solving Problems by Searching Blind Search, Informed Search and Local Search Multiple Agents Search Adversarial Search and Game Playing Knowledge Representation, Reasoning and Planning Logic-based Intelligent Systems and Reasoning Planning Other Knowledge Representation Languages Reasoning with Uncertainty

AIS - Intro p. 3

Course Overview
Introduction: AI concepts and Intelligent Agents Solving Problems by Searching Blind Search, Informed Search and Local Search Multiple Agents Search Adversarial Search and Game Playing Knowledge Representation, Reasoning and Planning Logic-based Intelligent Systems and Reasoning Planning Other Knowledge Representation Languages Reasoning with Uncertainty Learning
AIS - Intro p. 3

Course Evaluation
Assignments 30%

AIS - Intro p. 4

Course Evaluation
Assignments 30% 1 programming assignment

AIS - Intro p. 4

Course Evaluation
Assignments 30% 1 programming assignment 1 - 3 homeworks/quizzes

AIS - Intro p. 4

Course Evaluation
Assignments 30% 1 programming assignment 1 - 3 homeworks/quizzes class activities

AIS - Intro p. 4

Course Evaluation
Assignments 30% 1 programming assignment 1 - 3 homeworks/quizzes class activities MidTest 35%

AIS - Intro p. 4

Course Evaluation
Assignments 30% 1 programming assignment 1 - 3 homeworks/quizzes class activities MidTest 35% Final Test 35%

AIS - Intro p. 4

Textbooks
Russell & Norvig, Articial Intelligence, A Modern Approach, 3rd ed., Prentice-Hall, 2010

AIS - Intro p. 5

Textbooks
Russell & Norvig, Articial Intelligence, A Modern Approach, 3rd ed., Prentice-Hall, 2010 Poole & Mackworth, Articial Intelligence: Foundations of Computational Agents, Cambridge University Press, 2010

AIS - Intro p. 5

Textbooks
Russell & Norvig, Articial Intelligence, A Modern Approach, 3rd ed., Prentice-Hall, 2010 Poole & Mackworth, Articial Intelligence: Foundations of Computational Agents, Cambridge University Press, 2010 Nilsson, Articial Intelligence: A New Synthesis, Morgan Kaufmann, 1998

AIS - Intro p. 5

Textbooks
Russell & Norvig, Articial Intelligence, A Modern Approach, 3rd ed., Prentice-Hall, 2010 Poole & Mackworth, Articial Intelligence: Foundations of Computational Agents, Cambridge University Press, 2010 Nilsson, Articial Intelligence: A New Synthesis, Morgan Kaufmann, 1998 Andries P. Engelbrecht, Computational Intelligence An Introduction, 2nd ed., Wiley, 2007

AIS - Intro p. 5

Intro to AI
Backgrounds

AIS - Intro p. 6

Intro to AI
Backgrounds Paradigms of AI

AIS - Intro p. 6

Intro to AI
Backgrounds Paradigms of AI Foundations of AI

AIS - Intro p. 6

Intro to AI
Backgrounds Paradigms of AI Foundations of AI History of AI

AIS - Intro p. 6

Intro to AI
Backgrounds Paradigms of AI Foundations of AI History of AI State of the Art

AIS - Intro p. 6

Backgrounds - 1
Intelligence? What is that?

AIS - Intro p. 7

Backgrounds - 1
Intelligence? What is that? Articial? What is Articial Intelligence?

AIS - Intro p. 7

Backgrounds - 1
Intelligence? What is that? Articial? What is Articial Intelligence? How to develop intelligent systems?

AIS - Intro p. 7

Backgrounds - 1
Intelligence? What is that? Articial? What is Articial Intelligence? How to develop intelligent systems? encompasses a huge variety of subelds

AIS - Intro p. 7

Backgrounds - 1
Intelligence? What is that? Articial? What is Articial Intelligence? How to develop intelligent systems? encompasses a huge variety of subelds more basic question: Can machines think?

AIS - Intro p. 7

Backgrounds - 2
The word intelligence comes from inter (between) +legere (to choose)

AIS - Intro p. 8

Backgrounds - 2
The word intelligence comes from inter (between) +legere (to choose) We dene articial as "anything created/developed by humans"

AIS - Intro p. 8

Backgrounds - 2
The word intelligence comes from inter (between) +legere (to choose) We dene articial as "anything created/developed by humans" AI denitions are concerned with thought processes, reasoning and behaviours

AIS - Intro p. 8

Backgrounds - 2
The word intelligence comes from inter (between) +legere (to choose) We dene articial as "anything created/developed by humans" AI denitions are concerned with thought processes, reasoning and behaviours involves many other disciplines

AIS - Intro p. 8

Backgrounds - 2
The word intelligence comes from inter (between) +legere (to choose) We dene articial as "anything created/developed by humans" AI denitions are concerned with thought processes, reasoning and behaviours involves many other disciplines The term AI is gradually replaced by Intelligent Systems

AIS - Intro p. 8

AI Paradigms
thinking humanly

AIS - Intro p. 9

AI Paradigms
thinking humanly acting humanly

AIS - Intro p. 9

AI Paradigms
thinking humanly acting humanly thinking rationally

AIS - Intro p. 9

AI Paradigms
thinking humanly acting humanly thinking rationally acting rationally

AIS - Intro p. 9

Thinking Humanly
"The exciting new effort to make computers think. . . machines with minds, in the full and literal sense" [Haugeland, 1985]

AIS - Intro p. 10

Thinking Humanly
"The exciting new effort to make computers think. . . machines with minds, in the full and literal sense" [Haugeland, 1985] "(The automation of) activities that we associate with human thinking, activities such as decision-making, problem solving, learning . . . " [Bellman, 1978]

AIS - Intro p. 10

Thinking Humanly
"The exciting new effort to make computers think. . . machines with minds, in the full and literal sense" [Haugeland, 1985] "(The automation of) activities that we associate with human thinking, activities such as decision-making, problem solving, learning . . . " [Bellman, 1978] getting inside the actual workings of human minds

AIS - Intro p. 10

Thinking Humanly
"The exciting new effort to make computers think. . . machines with minds, in the full and literal sense" [Haugeland, 1985] "(The automation of) activities that we associate with human thinking, activities such as decision-making, problem solving, learning . . . " [Bellman, 1978] getting inside the actual workings of human minds developed theories of how human minds work

AIS - Intro p. 10

Thinking Humanly
"The exciting new effort to make computers think. . . machines with minds, in the full and literal sense" [Haugeland, 1985] "(The automation of) activities that we associate with human thinking, activities such as decision-making, problem solving, learning . . . " [Bellman, 1978] getting inside the actual workings of human minds developed theories of how human minds work Ex.: General Problem Solver [Newell & Simon, 1961], developed to compare machine reasoning with human reasoning

AIS - Intro p. 10

Thinking Humanly
"The exciting new effort to make computers think. . . machines with minds, in the full and literal sense" [Haugeland, 1985] "(The automation of) activities that we associate with human thinking, activities such as decision-making, problem solving, learning . . . " [Bellman, 1978] getting inside the actual workings of human minds developed theories of how human minds work Ex.: General Problem Solver [Newell & Simon, 1961], developed to compare machine reasoning with human reasoning is developed into cognitive science, which combines AI computer models and psychological experiments

AIS - Intro p. 10

Thinking Humanly
"The exciting new effort to make computers think. . . machines with minds, in the full and literal sense" [Haugeland, 1985] "(The automation of) activities that we associate with human thinking, activities such as decision-making, problem solving, learning . . . " [Bellman, 1978] getting inside the actual workings of human minds developed theories of how human minds work Ex.: General Problem Solver [Newell & Simon, 1961], developed to compare machine reasoning with human reasoning is developed into cognitive science, which combines AI computer models and psychological experiments still relevant in some AI elds, e.g. computer vision
AIS - Intro p. 10

Acting Humanly
"The art of creating machines that perform functions that require intelligence when performed by people" [Kurzweil, 1990]

AIS - Intro p. 11

Acting Humanly
"The art of creating machines that perform functions that require intelligence when performed by people" [Kurzweil, 1990] "The study of how to make computers do things at which, at the moment, people are better" [Rich & Knight, 1991]

AIS - Intro p. 11

Acting Humanly
"The art of creating machines that perform functions that require intelligence when performed by people" [Kurzweil, 1990] "The study of how to make computers do things at which, at the moment, people are better" [Rich & Knight, 1991] Alan Turing in 1950 proposed Turing Test

AIS - Intro p. 11

Acting Humanly
"The art of creating machines that perform functions that require intelligence when performed by people" [Kurzweil, 1990] "The study of how to make computers do things at which, at the moment, people are better" [Rich & Knight, 1991] Alan Turing in 1950 proposed Turing Test The goal is to test whether a computer program possess intelligence

AIS - Intro p. 11

Acting Humanly
"The art of creating machines that perform functions that require intelligence when performed by people" [Kurzweil, 1990] "The study of how to make computers do things at which, at the moment, people are better" [Rich & Knight, 1991] Alan Turing in 1950 proposed Turing Test The goal is to test whether a computer program possess intelligence Inspire the Captcha approach
AIS - Intro p. 11

Turing Test
Requires the computer to possess many capabilities: Natural Language Processing, Knowledge
Representation, Automated Reasoning, Machine Learning, Computer Vision, Robotics

AIS - Intro p. 12

Turing Test
Requires the computer to possess many capabilities: Natural Language Processing, Knowledge
Representation, Automated Reasoning, Machine Learning, Computer Vision, Robotics

Alan Turing listed some examples of things that machines can never do:Be kind, resourceful, beautiful,
friendly, have initiative, have a sense of humor, tell right from wrong, make mistakes, fall in love, enjoy strawberries and cream, make someone fall in love with it, learn from experience, use words properly, be the subject of its own thought, have as much diversity of behavior as man, do something really new.

AIS - Intro p. 12

Turing Test
Requires the computer to possess many capabilities: Natural Language Processing, Knowledge
Representation, Automated Reasoning, Machine Learning, Computer Vision, Robotics

Alan Turing listed some examples of things that machines can never do:Be kind, resourceful, beautiful,
friendly, have initiative, have a sense of humor, tell right from wrong, make mistakes, fall in love, enjoy strawberries and cream, make someone fall in love with it, learn from experience, use words properly, be the subject of its own thought, have as much diversity of behavior as man, do something really new.

It is realised later that some things in the list are possible to be done by machines.
AIS - Intro p. 12

Thinking Rationally
"The study of mental faculties through the use of computational models" [Charniak & McDermott, 1985]

AIS - Intro p. 13

Thinking Rationally
"The study of mental faculties through the use of computational models" [Charniak & McDermott, 1985] "The study of the computations that make it possible to perceive, reason, and act" [Winston, 1992]

AIS - Intro p. 13

Thinking Rationally
"The study of mental faculties through the use of computational models" [Charniak & McDermott, 1985] "The study of the computations that make it possible to perceive, reason, and act" [Winston, 1992] Started from the eld of Logic, initiated by Greek philosophers Aristotle

AIS - Intro p. 13

Thinking Rationally
"The study of mental faculties through the use of computational models" [Charniak & McDermott, 1985] "The study of the computations that make it possible to perceive, reason, and act" [Winston, 1992] Started from the eld of Logic, initiated by Greek philosophers Aristotle Rationality is dened in terms of logical rules

AIS - Intro p. 13

Thinking Rationally
"The study of mental faculties through the use of computational models" [Charniak & McDermott, 1985] "The study of the computations that make it possible to perceive, reason, and act" [Winston, 1992] Started from the eld of Logic, initiated by Greek philosophers Aristotle Rationality is dened in terms of logical rules Obstacles to this approach:

AIS - Intro p. 13

Thinking Rationally
"The study of mental faculties through the use of computational models" [Charniak & McDermott, 1985] "The study of the computations that make it possible to perceive, reason, and act" [Winston, 1992] Started from the eld of Logic, initiated by Greek philosophers Aristotle Rationality is dened in terms of logical rules Obstacles to this approach: Transformation from informal to formal knowledge is not always easy

AIS - Intro p. 13

Thinking Rationally
"The study of mental faculties through the use of computational models" [Charniak & McDermott, 1985] "The study of the computations that make it possible to perceive, reason, and act" [Winston, 1992] Started from the eld of Logic, initiated by Greek philosophers Aristotle Rationality is dened in terms of logical rules Obstacles to this approach: Transformation from informal to formal knowledge is not always easy Efciency as the result of expertise formed by experience (heuristics)
AIS - Intro p. 13

Acting Rationally
"the eld that studies the synthesis and analysis of computational agents that act intelligently" [Poole & Mackworth, 2010]

AIS - Intro p. 14

Acting Rationally
"the eld that studies the synthesis and analysis of computational agents that act intelligently" [Poole & Mackworth, 2010] "AI . . . is concerned with intelligent behavior in artifacts" [Nilsson, 1998]

AIS - Intro p. 14

Acting Rationally
"the eld that studies the synthesis and analysis of computational agents that act intelligently" [Poole & Mackworth, 2010] "AI . . . is concerned with intelligent behavior in artifacts" [Nilsson, 1998] based on the concept of rational agents

AIS - Intro p. 14

Acting Rationally
"the eld that studies the synthesis and analysis of computational agents that act intelligently" [Poole & Mackworth, 2010] "AI . . . is concerned with intelligent behavior in artifacts" [Nilsson, 1998] based on the concept of rational agents Rationality is dened as achieving the best outcome

AIS - Intro p. 14

Acting Rationally
"the eld that studies the synthesis and analysis of computational agents that act intelligently" [Poole & Mackworth, 2010] "AI . . . is concerned with intelligent behavior in artifacts" [Nilsson, 1998] based on the concept of rational agents Rationality is dened as achieving the best outcome Correct inference is only one of the possible mechanisms to achieve the best outcome

AIS - Intro p. 14

Acting Rationally
"the eld that studies the synthesis and analysis of computational agents that act intelligently" [Poole & Mackworth, 2010] "AI . . . is concerned with intelligent behavior in artifacts" [Nilsson, 1998] based on the concept of rational agents Rationality is dened as achieving the best outcome Correct inference is only one of the possible mechanisms to achieve the best outcome Efciency can be gained by acting based on limited rationality

AIS - Intro p. 14

Foundations of AI
philosophy: the basis of logical thinking

AIS - Intro p. 15

Foundations of AI
philosophy: the basis of logical thinking mathematics: formal logic, computation, probability

AIS - Intro p. 15

Foundations of AI
philosophy: the basis of logical thinking mathematics: formal logic, computation, probability economics: game theory, operations research

AIS - Intro p. 15

Foundations of AI
philosophy: the basis of logical thinking mathematics: formal logic, computation, probability economics: game theory, operations research neuroscience: ANN

AIS - Intro p. 15

Foundations of AI
philosophy: the basis of logical thinking mathematics: formal logic, computation, probability economics: game theory, operations research neuroscience: ANN psychology: cognitive science

AIS - Intro p. 15

Foundations of AI
philosophy: the basis of logical thinking mathematics: formal logic, computation, probability economics: game theory, operations research neuroscience: ANN psychology: cognitive science computer engineering: computer architecture

AIS - Intro p. 15

Foundations of AI
philosophy: the basis of logical thinking mathematics: formal logic, computation, probability economics: game theory, operations research neuroscience: ANN psychology: cognitive science computer engineering: computer architecture control theory: robotics

AIS - Intro p. 15

Foundations of AI
philosophy: the basis of logical thinking mathematics: formal logic, computation, probability economics: game theory, operations research neuroscience: ANN psychology: cognitive science computer engineering: computer architecture control theory: robotics linguistics: NLP
AIS - Intro p. 15

History of AI
gestation of AI: 1943 1955

AIS - Intro p. 16

History of AI
gestation of AI: 1943 1955 birth of AI: 1956

AIS - Intro p. 16

History of AI
gestation of AI: 1943 1955 birth of AI: 1956 early enthusiasm, great expectations: 1952 1969

AIS - Intro p. 16

History of AI
gestation of AI: 1943 1955 birth of AI: 1956 early enthusiasm, great expectations: 1952 1969 a dose of reality: 1966 1973

AIS - Intro p. 16

History of AI
gestation of AI: 1943 1955 birth of AI: 1956 early enthusiasm, great expectations: 1952 1969 a dose of reality: 1966 1973 knowledge based systems: 1969 1979

AIS - Intro p. 16

History of AI
gestation of AI: 1943 1955 birth of AI: 1956 early enthusiasm, great expectations: 1952 1969 a dose of reality: 1966 1973 knowledge based systems: 1969 1979 AI industry: 1980 present

AIS - Intro p. 16

History of AI
gestation of AI: 1943 1955 birth of AI: 1956 early enthusiasm, great expectations: 1952 1969 a dose of reality: 1966 1973 knowledge based systems: 1969 1979 AI industry: 1980 present the return of ANN: 1986 present

AIS - Intro p. 16

History of AI
gestation of AI: 1943 1955 birth of AI: 1956 early enthusiasm, great expectations: 1952 1969 a dose of reality: 1966 1973 knowledge based systems: 1969 1979 AI industry: 1980 present the return of ANN: 1986 present adoption of scientic methods: 1987 present

AIS - Intro p. 16

History of AI
gestation of AI: 1943 1955 birth of AI: 1956 early enthusiasm, great expectations: 1952 1969 a dose of reality: 1966 1973 knowledge based systems: 1969 1979 AI industry: 1980 present the return of ANN: 1986 present adoption of scientic methods: 1987 present intelligent agents: 1995 present

AIS - Intro p. 16

History of AI
gestation of AI: 1943 1955 birth of AI: 1956 early enthusiasm, great expectations: 1952 1969 a dose of reality: 1966 1973 knowledge based systems: 1969 1979 AI industry: 1980 present the return of ANN: 1986 present adoption of scientic methods: 1987 present intelligent agents: 1995 present large datasets: 2001 present
AIS - Intro p. 16

The Gestation of AI
McCulloch and Pitts: a model of articial neurons [1943]

AIS - Intro p. 17

The Gestation of AI
McCulloch and Pitts: a model of articial neurons [1943] Donald Hebb: Hebbian learning, learning by adjusting neuron connections [1949]

AIS - Intro p. 17

The Gestation of AI
McCulloch and Pitts: a model of articial neurons [1943] Donald Hebb: Hebbian learning, learning by adjusting neuron connections [1949] Minsky and Edmonds: rst neural network computer, SNARC [1950]

AIS - Intro p. 17

The Gestation of AI
McCulloch and Pitts: a model of articial neurons [1943] Donald Hebb: Hebbian learning, learning by adjusting neuron connections [1949] Minsky and Edmonds: rst neural network computer, SNARC [1950] Alan Turing: Computing Machinery and Intelligence, a visionary article [1947]

AIS - Intro p. 17

The birth of AI
First used in a two-month workshop in Dartmouth College

AIS - Intro p. 18

The birth of AI
First used in a two-month workshop in Dartmouth College initiated by John McCarthy

AIS - Intro p. 18

The birth of AI
First used in a two-month workshop in Dartmouth College initiated by John McCarthy attended by Marvin Minsky, Claude Shannon, Nathaniel Rochester

AIS - Intro p. 18

The birth of AI
First used in a two-month workshop in Dartmouth College initiated by John McCarthy attended by Marvin Minsky, Claude Shannon, Nathaniel Rochester also by Trenchard More, Arthur Samuel, Ray Solomonoff, and Oliver Selfridge

AIS - Intro p. 18

The birth of AI
First used in a two-month workshop in Dartmouth College initiated by John McCarthy attended by Marvin Minsky, Claude Shannon, Nathaniel Rochester also by Trenchard More, Arthur Samuel, Ray Solomonoff, and Oliver Selfridge and Allen Newell and Herbert Simon

AIS - Intro p. 18

Early Enthusiasm, Great Expectations

Newell & Simon: Logic Theorist, General Problem Solver, physical symbol system hypothesis

AIS - Intro p. 19

Early Enthusiasm, Great Expectations

Newell & Simon: Logic Theorist, General Problem Solver, physical symbol system hypothesis Shannons chess player and Samuels checkers player (includes learning)

AIS - Intro p. 19

Early Enthusiasm, Great Expectations

Newell & Simon: Logic Theorist, General Problem Solver, physical symbol system hypothesis Shannons chess player and Samuels checkers player (includes learning) Herbert Gelernter: Geometry Theorem Prover

AIS - Intro p. 19

Early Enthusiasm, Great Expectations

Newell & Simon: Logic Theorist, General Problem Solver, physical symbol system hypothesis Shannons chess player and Samuels checkers player (includes learning) Herbert Gelernter: Geometry Theorem Prover John McCarthy: LISP, time-sharing, Advice Taker (the rst hypothetical complete AI program)

AIS - Intro p. 19

Early Enthusiasm, Great Expectations

Newell & Simon: Logic Theorist, General Problem Solver, physical symbol system hypothesis Shannons chess player and Samuels checkers player (includes learning) Herbert Gelernter: Geometry Theorem Prover John McCarthy: LISP, time-sharing, Advice Taker (the rst hypothetical complete AI program) J. A. Robinson: resolution

AIS - Intro p. 19

Early Enthusiasm, Great Expectations

Newell & Simon: Logic Theorist, General Problem Solver, physical symbol system hypothesis Shannons chess player and Samuels checkers player (includes learning) Herbert Gelernter: Geometry Theorem Prover John McCarthy: LISP, time-sharing, Advice Taker (the rst hypothetical complete AI program) J. A. Robinson: resolution Fikes & Nilsson: SHAKEY robot

AIS - Intro p. 19

Early Enthusiasm, Great Expectations

Newell & Simon: Logic Theorist, General Problem Solver, physical symbol system hypothesis Shannons chess player and Samuels checkers player (includes learning) Herbert Gelernter: Geometry Theorem Prover John McCarthy: LISP, time-sharing, Advice Taker (the rst hypothetical complete AI program) J. A. Robinson: resolution Fikes & Nilsson: SHAKEY robot Marvin Minskys students: microworlds

AIS - Intro p. 19

Early Enthusiasm, Great Expectations

Newell & Simon: Logic Theorist, General Problem Solver, physical symbol system hypothesis Shannons chess player and Samuels checkers player (includes learning) Herbert Gelernter: Geometry Theorem Prover John McCarthy: LISP, time-sharing, Advice Taker (the rst hypothetical complete AI program) J. A. Robinson: resolution Fikes & Nilsson: SHAKEY robot Marvin Minskys students: microworlds Widrows adalines and Rosenblatts perceptrons
AIS - Intro p. 19

A Dose of Reality
Simons prediction (1957) a computer that becomes chess champion within 10 years

AIS - Intro p. 20

A Dose of Reality
Simons prediction (1957) a computer that becomes chess champion within 10 years It was achieved in 40 years!

AIS - Intro p. 20

A Dose of Reality
Simons prediction (1957) a computer that becomes chess champion within 10 years It was achieved in 40 years! AI programs need to know more about domain knowledge

AIS - Intro p. 20

A Dose of Reality
Simons prediction (1957) a computer that becomes chess champion within 10 years It was achieved in 40 years! AI programs need to know more about domain knowledge Computational complexity

AIS - Intro p. 20

A Dose of Reality
Simons prediction (1957) a computer that becomes chess champion within 10 years It was achieved in 40 years! AI programs need to know more about domain knowledge Computational complexity Support for AI research decreases

AIS - Intro p. 20

Knowledge-Based Systems
many general-purpose AI systems failed, on the other hand domain-specic AI systems gained some successes

AIS - Intro p. 21

Knowledge-Based Systems
many general-purpose AI systems failed, on the other hand domain-specic AI systems gained some successes DENDRAL: inferring molecular structure from mass [1969]

AIS - Intro p. 21

Knowledge-Based Systems
many general-purpose AI systems failed, on the other hand domain-specic AI systems gained some successes DENDRAL: inferring molecular structure from mass [1969] MYCIN: diagnosing blood infections

AIS - Intro p. 21

Knowledge-Based Systems
many general-purpose AI systems failed, on the other hand domain-specic AI systems gained some successes DENDRAL: inferring molecular structure from mass [1969] MYCIN: diagnosing blood infections Knowledge representation in NLP

AIS - Intro p. 21

Knowledge-Based Systems
many general-purpose AI systems failed, on the other hand domain-specic AI systems gained some successes DENDRAL: inferring molecular structure from mass [1969] MYCIN: diagnosing blood infections Knowledge representation in NLP PROLOG and frames as knowledge representation

AIS - Intro p. 21

AI Industry
R1: the rst commercial expert system in DEC

AIS - Intro p. 22

AI Industry
R1: the rst commercial expert system in DEC By 1988, DEC had deployed 40 expert systems

AIS - Intro p. 22

AI Industry
R1: the rst commercial expert system in DEC By 1988, DEC had deployed 40 expert systems Japan, USA and UK developed ambitious AI projects

AIS - Intro p. 22

AI Industry
R1: the rst commercial expert system in DEC By 1988, DEC had deployed 40 expert systems Japan, USA and UK developed ambitious AI projects In the end, they failed to meet their goals

AIS - Intro p. 22

AI Industry
R1: the rst commercial expert system in DEC By 1988, DEC had deployed 40 expert systems Japan, USA and UK developed ambitious AI projects In the end, they failed to meet their goals AI winter

AIS - Intro p. 22

The Return of ANN

reinvention of backpropagation

AIS - Intro p. 23

The Return of ANN

reinvention of backpropagation Connectionist model was seen as competitor of symbolic model and also logicist model

AIS - Intro p. 23

The Adoption of Scientic Methods

Built on existing theories

AIS - Intro p. 24

The Adoption of Scientic Methods

Built on existing theories Based claims on rigorous theorems or experimental results

AIS - Intro p. 24

The Adoption of Scientic Methods

Built on existing theories Based claims on rigorous theorems or experimental results HMM dominated speech recognition

AIS - Intro p. 24

The Adoption of Scientic Methods

Built on existing theories Based claims on rigorous theorems or experimental results HMM dominated speech recognition Data mining technology

AIS - Intro p. 24

The Adoption of Scientic Methods

Built on existing theories Based claims on rigorous theorems or experimental results HMM dominated speech recognition Data mining technology Bayesian network for probabilistic reasoning

AIS - Intro p. 24

Intelligent Agents
SOAR: complete agent architecture

AIS - Intro p. 25

Intelligent Agents
SOAR: complete agent architecture Many intelligent agents in internet

AIS - Intro p. 25

Intelligent Agents
SOAR: complete agent architecture Many intelligent agents in internet Human Level AI, not only specic tasks

AIS - Intro p. 25

Intelligent Agents
SOAR: complete agent architecture Many intelligent agents in internet Human Level AI, not only specic tasks Articial General Intelligence, which looks for a universal algorithm for learning and acting in any environment

AIS - Intro p. 25

The Availability of Large Datasets

Intelligence based on large corpus in NLP

AIS - Intro p. 26

The Availability of Large Datasets

Intelligence based on large corpus in NLP Also in image processing

AIS - Intro p. 26

State of The Art

Robotic vehicle called STANLEY (tested on dessert) and BOSS (tested in closed area for safety driving)

AIS - Intro p. 27

State of The Art

Robotic vehicle called STANLEY (tested on dessert) and BOSS (tested in closed area for safety driving) Speech recognition and dialog management system in United Airlines

AIS - Intro p. 27

State of The Art

Robotic vehicle called STANLEY (tested on dessert) and BOSS (tested in closed area for safety driving) Speech recognition and dialog management system in United Airlines NASAs Remote Agent: planning and scheduling (and monitoring) operations of spacecraft

AIS - Intro p. 27

State of The Art

Robotic vehicle called STANLEY (tested on dessert) and BOSS (tested in closed area for safety driving) Speech recognition and dialog management system in United Airlines NASAs Remote Agent: planning and scheduling (and monitoring) operations of spacecraft IBMs Deep Blue defeated Garry Kasparov

AIS - Intro p. 27

State of The Art

Robotic vehicle called STANLEY (tested on dessert) and BOSS (tested in closed area for safety driving) Speech recognition and dialog management system in United Airlines NASAs Remote Agent: planning and scheduling (and monitoring) operations of spacecraft IBMs Deep Blue defeated Garry Kasparov Learning algorithms for spam ghting

AIS - Intro p. 27

State of The Art

Robotic vehicle called STANLEY (tested on dessert) and BOSS (tested in closed area for safety driving) Speech recognition and dialog management system in United Airlines NASAs Remote Agent: planning and scheduling (and monitoring) operations of spacecraft IBMs Deep Blue defeated Garry Kasparov Learning algorithms for spam ghting DART program performed planning in Gulf War

AIS - Intro p. 27

State of The Art

Robotic vehicle called STANLEY (tested on dessert) and BOSS (tested in closed area for safety driving) Speech recognition and dialog management system in United Airlines NASAs Remote Agent: planning and scheduling (and monitoring) operations of spacecraft IBMs Deep Blue defeated Garry Kasparov Learning algorithms for spam ghting DART program performed planning in Gulf War Over 2 million Roomba vacuum cleaner robots

AIS - Intro p. 27

State of The Art

Robotic vehicle called STANLEY (tested on dessert) and BOSS (tested in closed area for safety driving) Speech recognition and dialog management system in United Airlines NASAs Remote Agent: planning and scheduling (and monitoring) operations of spacecraft IBMs Deep Blue defeated Garry Kasparov Learning algorithms for spam ghting DART program performed planning in Gulf War Over 2 million Roomba vacuum cleaner robots A program translates Arabic to English
AIS - Intro p. 27

Conclusion and Future Works

A shift of paradigm: from AI to IS

AIS - Intro p. 28

Conclusion and Future Works

A shift of paradigm: from AI to IS Systematic approach to learn IS

AIS - Intro p. 28

Conclusion and Future Works

A shift of paradigm: from AI to IS Systematic approach to learn IS Common components of IS

AIS - Intro p. 28

Conclusion and Future Works

A shift of paradigm: from AI to IS Systematic approach to learn IS Common components of IS Relation between IS and Computer Science

AIS - Intro p. 28