Introduction to How does How do we Artificial Intelligence the - - PDF document

1

Introduction to Artificial Intelligence

—————————————— Christian Jacob

jacob@cpsc.ucalgary.ca

Department of Computer Science University of Calgary

1. What is “Artificial Intelligence”?

How does the human brain work? How do we emulate the human brain? Who cares? Let’s do some cool and useful stuff! How do we create intelligence? What is intelligence?

2. Basic Problem-Solving Strategies

✦ Basic search techniques ✦ Problem decompositon and AND/OR

graphs

✦ Searching with problem-specific

knowledge

2. Basic Problem-Solving Strategies

✦ Basic search techniques ✦ Problem decompositon and AND/OR

graphs

✦ Searching with problem-specific

knowledge

2.1 Basic Search Techniques

✦ Greedy search ✦ Gradient descent or ascent ✦ Stochastic search

– Simulated annealing – Evolutionary search

✦ Depth-first vs. breadth-first search local maxima local maxima

In search for the highest landmark ...

global maximum

2 2. Basic Problem-Solving Strategies

✦ Basic search techniques ✦ Problem decompositon and AND/OR

graphs

✦ Searching with problem-specific

knowledge

2.2 Problem Decomposition and AND/OR Graphs

[Bratko, 2001] [Bratko, 2001]

2. Basic Problem-Solving Strategies

✦ Basic search techniques ✦ Problem decompositon and AND/OR

graphs

✦ Searching with problem-specific

knowledge

2.3 Searching with Problem-Specific Knowledge

75 118 111 140 80 97 99 101 211 A B C D E F G H I

State h(n) A 366 B 374 C 329 D 244 E 253 F 178 G 193 H 98 I

A E B C h=253 h=329 h=374

2.3 Searching with Problem-Specific Knowledge (2)

[Newborn, 1997] [Kurzweil, 1990] Anatoly Karpow and Gary Kasparow, 1986

3. Knowledge Representation, Reasoning, and Planning

✦ Knowledge representation ✦ Reasoning and planning ✦ Knowledge soup

3 3. Knowledge Representation, Reasoning, and Planning

✦ Knowledge representation ✦ Reasoning and planning ✦ Knowledge soup

3.1 Knowledge Representation

[Sowa, 2000]

3. Knowledge Representation, Reasoning, and Planning

✦ Knowledge representation ✦ Reasoning and planning ✦ Knowledge soup

3.2 Reasoning and Planning

Start Finish

At(Home), Rents(Rogers, Video), Sells(Store, Bread) At(Home), Have(Video), Have(Bread)

Rent(Video) Buy(Bread)

At(Rogers), Rents(Rogers, Video) At(Store), Sells(Store, Bread)

Go(Rogers) Go(Store)

At(Home) At(Home)

3. Knowledge Representation, Reasoning, and Planning

✦ Knowledge representation ✦ Reasoning and planning ✦ Knowledge soup

3.3 Knowledge Soup

✦ Vagueness, uncertainty, randomness,

ignorance

[Sowa, 2000]

4 4. Machine Learning and Pattern Recognition

✦ Fuzzy logic and fuzzy sets ✦ Artificial neural networks

– Single- and multilayer perceptrons – Backpropagation networks – Self-organizing feature maps – Recurrent networks

✦ Neuro-fuzzy systems

4. Machine Learning and Pattern Recognition

✦ Fuzzy logic and fuzzy sets ✦ Artificial neural networks

– Single- and multilayer perceptrons – Backpropagation networks – Self-organizing feature maps – Recurrent networks

✦ Neuro-fuzzy systems

4.1 Fuzziness

✦ What is hot? What is cold? ✦ What is young? What is old?

[Sowa, 2000]

4. Machine Learning and Pattern Recognition

✦ Fuzzy logic and fuzzy sets ✦ Artificial neural networks

– Single- and multilayer perceptrons – Backpropagation networks – Self-organizing feature maps – Recurrent networks

✦ Neuro-fuzzy systems

4.2 Artificial Neural Networks: Modeling the Brain

[Stevens et al., 1988]

Visual Cortex of a Cat

4.2 Artificial Neural Networks (2)

[Kurzweil, 1990] [Spektrum, 1993]

Schematic Perceptron Feed-forward network

5 Modeling the Brain?

[Spektrum, 1993] [Kurzweil, 1990]

Optical Character Recognition

[Spektrum, 1993]

Input

Hidden Output

4. Machine Learning and Pattern Recognition

✦ Fuzzy logic and fuzzy sets ✦ Artificial neural networks

– Single- and multilayer perceptrons – Backpropagation networks – Self-organizing feature maps – Recurrent networks

✦ Neuro-fuzzy systems

5. Evolutionary Computing

✦ Evolution Strategies ✦ Evolutionary Programming ✦ Genetic Algorithms ✦ Genetic Programming ✦ Learning Classifier Systems ✦ Evolutionary Design

5. Evolutionary Computing

✦ Evolution Strategies ✦ Evolutionary Programming ✦ Genetic Algorithms ✦ Genetic Programming ✦ Learning Classifier Systems ✦ Evolutionary Design

5.1 Evolution Strategies

[Rechenberg, 1994]

Evolution of a Jet Nozzle Evolutionary Engineering

6 5. Evolutionary Computing

✦ Evolution Strategies ✦ Evolutionary Programming ✦ Genetic Algorithms ✦ Genetic Programming ✦ Learning Classifier Systems ✦ Evolutionary Design

5.2 Evolutionary Programming

A0 A0 output Input A1 A2 A3 A1 output A2 output A3 output

Evolution

• f

Finite State Automata

[Jacob, 2001]

5.2 Evolutionary Programming

• Gen. 0
• Gen. 40
• Gen. 70
• Gen. 0
• Gen. 40
• Gen. 70
• Gen. 120

[Jacob, 2001]

5. Evolutionary Computing

✦ Evolution Strategies ✦ Evolutionary Programming ✦ Genetic Algorithms ✦ Genetic Programming ✦ Learning Classifier Systems ✦ Evolutionary Design

5.3 Genetic Algorithms

Genotype

{1,0,1,1,0,1,0,0,1,0,1,1} {0,1,1,1,1,0,0,1,0,0,0,1} {0,0,1,1,0,101,1,0,1,0,0} ... {1,1,0,0,0,1,0,1,0,1,0,0} ... {1,0,1,0,0,1,1,1,0,1,1,1} {0,0,1,1,0,1,1,1,0,1,0,0} {1,0,0,1,0,1,1,1,0,0,0,1}

Binary Vector Decoding Interpretation

Phenotype

5.3 Genetic Algorithms

Generation 0 Generation 10 Generation 11 Generation 30

[Jacob, 2001]

7 5. Evolutionary Computing

✦ Evolution Strategies ✦ Evolutionary Programming ✦ Genetic Algorithms ✦ Genetic Programming ✦ Learning Classifier Systems ✦ Evolutionary Design

5.4 Genetic Programming

sprout[3] Axiom [] LEFT F[_] PRED B[.38] leaf[0] sprout[2] F[1.4] YL[30] leaf[0] Seq bloom[1] Stack Seq RL[20] F[4.9] Stack SUCC [] RIGHT LRule ... LRule LRules LSystem sprout[3] Axiom [] LEFT _PRED [] RIGHT _SUCC LRule sproutIndex sprout PRED _SEQ _STACK Alternative SUCC _sprout _stalk ... _YL _SEQ Alternative BlankSequence SEQ __LRule LRules _Axiom _LRules LSystem AntTracker[ seq[seq[advance], ifSensor[dust][seq[seq[stop]], seq[advance]], seq[ ifSensor[dust][seq[turnLeft, again], seq[nop]], seq[seq[advance], ifSensor[wall][ seq[seq[stop], seq[nop], ifSensor[phero][seq[turnLeft], seq[stop]]], seq[turnRight]]], seq[nop]] ]]

• Gen. 5
• Gen. 16

AntTracker[ seq[ ifSensor[wall][ seq[turnLeft, again], seq[turnRight]], seq[ ifSensor[dust][seq[turnLeft, again],seq[nop]], seq[ ifSensor[wall][ seq[seq[turnLeft, again], seq[nop], ifSensor[phero][seq[stop]]], seq[seq[advance]]]], seq[nop]] ]]

• Gen. 22

AntTracker[ seq[ ifSensor[wall][ seq[turnLeft, turnLeft, again], seq[ ifSensor[wall][seq[turnLeft, again], seq[turnRight]], seq[ seq[ ifSensor[wall][ seq[seq[turnLeft, again], seq[nop], ifSensor[wall][seq[again], seq[seq[nop], seq[nop]]]], seq[seq[advance]]]]]]], seq[ ifSensor[dust][seq[turnLeft, again]], seq[ ifSensor[wall][seq[ifSensor[dust][seq[advance]]], seq[ifSensor[dust][seq[turnLeft, again]], seq[ifSensor[wall][ seq[seq[turnLeft, again], seq[nop], ifSensor[dust][seq[stop], seq[stop], seq[turnLeft]]], seq[seq[advance]]]], seq[nop]]] ]]]]

• Gen. 59
• Gen. 1

AntTracker[ seq[advance] ] AntTracker[ seq[seq[advance], ifSensor[wall][ seq[seq[stop], seq[nop], ifSensor[phero][seq[turnLeft], seq[advance]]], seq[turnRight]], seq[ifSensor[dust][seq[turnLeft, again],seq[nop]], seq[ifSensor[wall][ seq[seq[turnLeft, again], seq[nop], ifSensor[phero][seq[stop]]], seq[seq[advance]]]], seq[nop]]]]

Fitness Generation

59 45

AntTracker[ seq[ ifSensor[wall][ seq[turnLeft, turnLeft, again], seq[ ifSensor[wall][seq[turnLeft, again], seq[turnRight]], seq[ seq[ ifSensor[wall][ seq[seq[turnLeft, again], seq[nop], ifSensor[wall][seq[again], seq[seq[nop], seq[nop]]]], seq[seq[advance]]]]]]], seq[ ifSensor[dust][seq[turnLeft, again]], seq[ ifSensor[wall][seq[ifSensor[dust][ seq[advance]]], seq[ifSensor[dust][seq[turnLeft, again]], seq[ifSensor[wall][ seq[seq[turnLeft, again], seq[nop], ifSensor[dust][seq[stop], seq[stop], seq[turnLeft]]], seq[seq[advance]]]], seq[nop]]] ]]]]

[Jacob, 2001]

5. Evolutionary Computing

✦ Evolution Strategies ✦ Evolutionary Programming ✦ Genetic Algorithms ✦ Genetic Programming ✦ Learning Classifier Systems ✦ Evolutionary Design

5.5 Learning Classifier Systems

✦ A classifier system to emulate a frog.

The frog reacts to objects it sees.

Moving On the Ground Large Far Striped Flee! Pursue!

Input: Output:

1 _ _ _ _ 1 1 _ 1 1 1

8 5. Evolutionary Computing

✦ Evolution Strategies ✦ Evolutionary Programming ✦ Genetic Algorithms ✦ Genetic Programming ✦ Learning Classifier Systems ✦ Evolutionary Design

5.6 Evolutionary Design: Objects

Scaffold Lifting Loads

[Funes and Pollack, 1999]

5.6 Evolutionary Design: Art

Mutations

Hölldobler & Wilson, 1990

6. Swarm Intelligence and Complex Adaptive Systems

✦ Social Models ✦ Swarms and Emergent System

Behaviour

✦ Immune System Computing

6. Swarm Intelligence and Complex Adaptive Systems

✦ Social Models ✦ Swarms and Emergent System

Behaviour

✦ Immune System Computing

6.1 Social Models: Competition and Cooperation

[Ernst, 1998] [Nuridsany & Pérennou, 1996]

9 6. Swarm Intelligence and Complex Adaptive Systems

✦ Social Models ✦ Swarms and Emergent System

Behaviour

✦ Immune System Computing

6.2 Swarms: The Ants Paradigm

[Hölldobler & Wilson, 1990]

6.2 Swarms and Emergent System Behaviour 6. Swarm Intelligence and Complex Adaptive Systems

✦ Social Models ✦ Swarms and Emergent System

Behaviour

✦ Immune System Computing

6.3 Immune System Computing

✦ Distinguishing self from non-self

[Hofmeyr and Forrest, 1999]

Intrusion detection in a LAN

7. Robo Sapiens?

[Kurzweil, 1990]

Seymour Papert

LOGO Robot

10 7. Robo Sapiens?

[Menzel and D’Aluisio, 2000]

KISMET MIT

7. Robo Sapiens?

[Kurzweil, 1990]

WABOT, theOrgan Player

Ichiro Kato, Waseda-University, Tokyo

Artificial Intelligence in Action

If we don’t know how it works, then it’s AI. When we find out how it works, it’s not AI anymore…

References

• Bratko, I. (2001). PROLOG Programming for

Artificial Intelligence. New York, Addison- Wesley.

• Rechenberg, I. (1994). Evolutionsstrategie’94.

Stuttgart, Frommann-Holzboog.

• Sowa, J. F. (2000). Knowledge
• Representation. Pacific Grove, Brooks/Cole.
• Kurzweil, R. (1990). The Age of Intelligent
• Machines. Cambridge, MA, MIT Press.

References (2)

• P. Menzel and F. D’Aluisio (2000). Robo

sapiens — Evolution of a New Species. Cambridge, MA, MIT Press.

• Spektrum der Wissenschaft: Spezial. Gehirn

und Geist. Heidelberg, Spektrum Akademischer Verlag,1993.

• Hofmeyr, S. and Forrest, S. (1999). Immunity

by Design: An Artificial Immune System. In GECCO’99.

• Stevens, C. F., et al. (1988). Gehirn und
• Nervensystem. Heidelberg, Spektrum

Akademischer Verlag.

11 References (3)

• Ernst, A. M., ed. (1998). Digest: Kooperation

und Konkurrenz, Heidelberg, Spektrum Akademischer Verlag.

• Nuridsany, C., and Pérennou, M. (1996).

Microcosmos: The Invisible World of Insects. New York, Stewart, Tabori & Chang.

• Newborn, M. (1997). Kasparov versus Deep
• Blue. Berlin, Springer-Verlag.
• Jacob, C. (2001). Illustrating Evolutionary

Computation with Mathematica. San Francisco, Morgan Kaufmann.

References (4)

✦ Hölldobler, B., and Wilson, E. O. (1990). The Ants.

Cambridge, MA, Harvard University Press.

✦ Todd, S. and Latham, W. (1992). Evolutionary Art

and Computers. London, Academic Press.

✦ Funes, P. and Pollack, J. (1999). Computer

Evolution of Buildable Objects. In: P. Bentley (ed.). Evolutionary Design by Computers. San Francisco, Morgan Kaufmann.