+612 9045 4394
Anticipatory Learning Classifier Systems : Genetic Algorithms and Evolutionary Computation - Martin V. Butz

Anticipatory Learning Classifier Systems

Genetic Algorithms and Evolutionary Computation

Hardcover Published: 31st January 2002
ISBN: 9780792376309
Number Of Pages: 172

Share This Book:


RRP $514.99
or 4 easy payments of $89.13 with Learn more
Ships in 7 to 10 business days

Other Available Editions (Hide)

  • Paperback View Product Published: 5th November 2012

Anticipatory Learning Classifier Systems describes the state of the art of anticipatory learning classifier systems-adaptive rule learning systems that autonomously build anticipatory environmental models. An anticipatory model specifies all possible action-effects in an environment with respect to given situations. It can be used to simulate anticipatory adaptive behavior.

Anticipatory Learning Classifier Systems highlights how anticipations influence cognitive systems and illustrates the use of anticipations for (1) faster reactivity, (2) adaptive behavior beyond reinforcement learning, (3) attentional mechanisms, (4) simulation of other agents and (5) the implementation of a motivational module. The book focuses on a particular evolutionary model learning mechanism, a combination of a directed specializing mechanism and a genetic generalizing mechanism. Experiments show that anticipatory adaptive behavior can be simulated by exploiting the evolving anticipatory model for even faster model learning, planning applications, and adaptive behavior beyond reinforcement learning.

Anticipatory Learning Classifier Systems gives a detailed algorithmic description as well as a program documentation of a C++ implementation of the system. It is an excellent reference for researchers interested in adaptive behavior and machine learning from a cognitive science perspective as well as those who are interested in combining evolutionary learning mechanisms for learning and optimization tasks.

List of Figuresp. ix
List of Tablesp. xvi
Forewordp. xvii
Prefacep. xix
Complex Systems Approachp. xx
Towards ACS2p. xxiii
ACS2p. xxiv
Road Mapp. xxv
Acknowledgmentsp. xxvii
Backgroundp. 1
Anticipationsp. 2
Psychology Discovers Anticipationsp. 2
Theory of Anticipatory Behavioral Controlp. 3
Importance of Anticipationsp. 4
Genetic Algorithmsp. 6
Evolutionary Principlesp. 6
GA Frameworkp. 8
An Illustrative Examplep. 10
Learning Classifier Systemsp. 11
Holland's Cognitive Systemp. 13
LCS frameworkp. 14
Problems in Traditional LCSsp. 15
XCS Classifier Systemp. 16
ACS2p. 23
Frameworkp. 25
Environmental Interactionp. 25
Knowledge Representationp. 26
A Behavioral Actp. 27
Reinforcement Learningp. 29
The Anticipatory Learning Processp. 30
The Process in Detailp. 30
The ALP in Action: A Simple Gripper Problemp. 33
Causes for Over-Specializationp. 35
Genetic Generalization in ACS2p. 37
Accurate, Maximally General Classifiers in ACS2p. 38
The GA Ideap. 39
How the GA Worksp. 41
Interaction of ALP, GA, RL, and Behaviorp. 43
Subsumptionp. 44
Evolutionary Pressures of ALP and GAp. 45
All Interactionsp. 47
Experiments with ACS2p. 51
Gripper Problem Revisitedp. 52
Population without GAp. 52
Population with GAp. 54
Multiplexer Problemp. 55
Environmental Settingp. 56
Evolution of a Multiplexer Modelp. 57
ACS2 as a Classifierp. 63
Maze Environmentp. 64
Environmental Settingp. 65
Maze6p. 66
Woods14p. 68
Blocks Worldp. 69
Environmental Settingp. 71
Model Learningp. 73
Hand-Eye Coordination Taskp. 76
Environmental Settingp. 76
Model Learningp. 78
Result Summaryp. 79
Limitsp. 81
GA Challengesp. 81
Overlapping Classifiersp. 82
Interfering Specificitiesp. 85
Non-determinism and a First Approachp. 87
ACS2 in a Non-determinism Taskp. 88
Probability-Enhanced Effectsp. 89
Model Aliasingp. 93
Model Exploitationp. 99
Improving Model Learningp. 99
Increasing Explorationp. 100
Combining Exploration with Action Planningp. 104
Enhancing Reinforcement Learningp. 107
Response-Effect Learning Taskp. 107
Mental Actingp. 108
Lookahead Action Selectionp. 110
ACS2 in the Response-Effect Taskp. 111
Stimulus-Response-Effect Taskp. 112
Model Exploitation Recapitulationp. 113
Related Systemsp. 115
Estimated Learning Algorithmp. 115
Dynap. 117
Schema Mechanismp. 118
Expectancy Model SRS/Ep. 119
Summary, Conclusions, and Future Workp. 121
Summaryp. 121
Model Representation Enhancementsp. 123
Classifier Structurep. 123
ACS2 Structurep. 126
Model Learning Modificationsp. 127
Observations in Naturep. 127
Relevance and Influencep. 130
Attentional Mechanismsp. 131
Additional Memoryp. 133
Adaptive Behaviorp. 134
Reinforcement Learning Processesp. 135
Behavioral Modulep. 136
ACS2 in the Futurep. 137
Appendicesp. 139
Parameters in ACS2p. 139
Algorithmic Description of ACS2p. 141
Initializationp. 141
The Main Execution Loopp. 142
Formation of the Match Setp. 143
Choosing an Actionp. 143
Formation of the Action Setp. 144
Application of the ALPp. 144
Reinforcement Learningp. 149
GA Applicationp. 149
Subsumptionp. 152
ACS2 C++ Code Documentationp. 153
Getting Startedp. 153
Structure of the Codep. 154
The Controller - ACSConstants.hp. 154
The Executer - acs2++.ccp. 156
Environmentsp. 157
ACS2 modulesp. 159
Performance Outputp. 160
Glossaryp. 161
Referencesp. 165
Indexp. 171
Table of Contents provided by Syndetics. All Rights Reserved.

ISBN: 9780792376309
ISBN-10: 0792376307
Series: Genetic Algorithms and Evolutionary Computation
Audience: Professional
Format: Hardcover
Language: English
Number Of Pages: 172
Published: 31st January 2002
Publisher: Springer
Country of Publication: NL
Dimensions (cm): 23.5 x 15.5  x 1.91
Weight (kg): 1.03