Springer
Table of ContentsAuthor IndexSearch

Tournament Selection: Stable Fitness Pressure in XCS

Martin V. Butz, Kumara Sastry, and David E. Goldberg

Illinois Genetic Algorithms Laboratory (IlliGAL)
University of Illinois at Urbana-Champaign,
104 S. Mathews
61801 Urbana, IL, USA
{butz,kumara,deg}@illigal.ge.uiuc.edu

Abstract. Although it is known from GA literature that proportionate selection is subject to many pitfalls, the LCS community somewhat adhered to proportionate selection. Also in the accuracy-based learning classifier system XCS, introduced by Wilson in 1995, proportionate selection is used. This paper identifies problem properties in which performance of proportionate selection is impaired. Consequently, tournament selection is introduced which makes XCS more parameter independent, noise independent, and more efficient in exploiting fitness guidance.

LNCS 2724, p. 1857 ff.

Full article in PDF


lncs@springer.de
© Springer-Verlag Berlin Heidelberg 2003