Many successful applications have proven the potential of Learning
Classifier Systems and the XCS classifier system in particular in
datamining, reinforcement learning, and function approximation tasks.
Recent research has shown that XCS is a highly flexible system, which
can be adapted to the task at hand by adjusting its condition structures,
learning operators, and prediction mechanisms. However, grounding
theory concerning the scalability of XCS dependent on these enhancements
and problem difficulty is still rather sparse and mainly restricted
to boolean function problems. In this article we developed a learning
scalability theory for XCSF---the XCS system applied to real-valued
function approximation problems. We determine crucial dependencies
on functional properties and on the developed solution representation
and derive a theoretical scalability model out of these constraints.
The theoretical model is verified with empirical evidence. That is,
we show that given a particular problem difficulty and particular
representational constraints XCSF scales optimally. In consequence,
we discuss the importance of appropriate prediction and condition
structures regarding a given problem and show that scalability properties
can be improved by polynomial orders, given an appropriate, problem-suitable
representation.
%0 Journal Article
%1 Stalph:2009
%A Stalph, P. O.
%A Butz, Martin V.
%A Goldberg, David E.
%A Llorà, Xavier
%D 2009
%J Genetic and Evolutionary Computation Conference, GECCO 2009
%K imported
%P in press
%T On the Scalability of XCS(F)
%X Many successful applications have proven the potential of Learning
Classifier Systems and the XCS classifier system in particular in
datamining, reinforcement learning, and function approximation tasks.
Recent research has shown that XCS is a highly flexible system, which
can be adapted to the task at hand by adjusting its condition structures,
learning operators, and prediction mechanisms. However, grounding
theory concerning the scalability of XCS dependent on these enhancements
and problem difficulty is still rather sparse and mainly restricted
to boolean function problems. In this article we developed a learning
scalability theory for XCSF---the XCS system applied to real-valued
function approximation problems. We determine crucial dependencies
on functional properties and on the developed solution representation
and derive a theoretical scalability model out of these constraints.
The theoretical model is verified with empirical evidence. That is,
we show that given a particular problem difficulty and particular
representational constraints XCSF scales optimally. In consequence,
we discuss the importance of appropriate prediction and condition
structures regarding a given problem and show that scalability properties
can be improved by polynomial orders, given an appropriate, problem-suitable
representation.
@article{Stalph:2009,
abstract = {Many successful applications have proven the potential of Learning
Classifier Systems and the XCS classifier system in particular in
datamining, reinforcement learning, and function approximation tasks.
Recent research has shown that XCS is a highly flexible system, which
can be adapted to the task at hand by adjusting its condition structures,
learning operators, and prediction mechanisms. However, grounding
theory concerning the scalability of XCS dependent on these enhancements
and problem difficulty is still rather sparse and mainly restricted
to boolean function problems. In this article we developed a learning
scalability theory for XCSF---the XCS system applied to real-valued
function approximation problems. We determine crucial dependencies
on functional properties and on the developed solution representation
and derive a theoretical scalability model out of these constraints.
The theoretical model is verified with empirical evidence. That is,
we show that given a particular problem difficulty and particular
representational constraints XCSF scales optimally. In consequence,
we discuss the importance of appropriate prediction and condition
structures regarding a given problem and show that scalability properties
can be improved by polynomial orders, given an appropriate, problem-suitable
representation.},
added-at = {2009-06-26T15:25:19.000+0200},
author = {Stalph, P. O. and Butz, Martin V. and Goldberg, David E. and Llor\`a, Xavier},
biburl = {https://www.bibsonomy.org/bibtex/23d0a365b17d2ce0074d9a30d6dc224d4/butz},
description = {diverse cognitive systems bib},
interhash = {7c6a91b14e89c2b071f0302b0d57d259},
intrahash = {3d0a365b17d2ce0074d9a30d6dc224d4},
journal = {Genetic and Evolutionary Computation Conference, {GECCO} 2009},
keywords = {imported},
owner = {butz},
pages = {in press},
timestamp = {2009-06-26T15:25:56.000+0200},
title = {On the Scalability of {XCS(F)}},
year = 2009
}