Butz Anticipatory Learning Classifier Systems

Erscheinungsjahr 2012
ISBN: 978-1-4615-0891-5
Verlag: Springer US
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, Band 4, 172 Seiten, eBook

Reihe: Genetic Algorithms and Evolutionary Computation

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.

Zielgruppe

Research

Autoren/Hrsg.

Butz, Martin V.

Inhaltsverzeichnis

1. Background.- 1. Anticipations.- 2. Genetic Algorithms.- 3. Learning Classifier Systems.- 2. ACS2.- 1. Framework.- 2. Reinforcement Learning.- 3. The Anticipatory Learning Process.- 4. Genetic Generalization in ACS2.- 5. Interaction of ALP, GA, RL, and Behavior.- 3. Experiments with ACS2.- 1. Gripper Problem Revisited.- 2. Multiplexer Problem.- 3. Maze Environment.- 4. Blocks World.- 5. Hand-Eye Coordination Task.- 6. Result Summary.- 4. Limits.- 1.GA Challenges.- 2.Non-determinism and a First Approach.- 3. Model Aliasing.- 5. Model Exploitation.- 1. Improving Model Learning.- 2. Enhancing Reinforcement Learning.- 3. Model Exploitation Recapitulation.- 6. Related Systems.- 1. Estimated Learning Algorithm.- 2. Dyna.- 3. Schema Mechanism.- 4. Expectancy Model SRS/E.- 7. Summary, Conclusions, and Future Work.- 1. Summary.- 2. Model Representation Enhancements.- 3. Model Learning Modifications.- 4. Adaptive Behavior.- 5. ACS2 in the Future.- Appendices.- References.