Introducing Subchromosome Representations to the Linkage Learning Genetic Algorithm

Ying-ping Chen¹ and David E. Goldberg²

¹Department of Computer Science and Department of General Engineering, University of Illinois, Urbana, IL 61801, USA
ypchen@illigal.ge.uiuc.edu

²Department of General Engineering, University of Illinois, Urbana, IL 61801, USA
deg@uiuc.edu

Abstract. This paper introduces subchromosome representations to the linkage learning genetic algorithm (LLGA). The subchromosome representation is utilized for effectively lowering the number of building blocks in order to escape from the performance limit implied by the convergence time model for the linkage learning genetic algorithm. A preliminary implementation to realize subchromosome representations is developed and tested. The experimental results indicate that the proposed representation can improve the performance of the linkage learning genetic algorithm on uniformly scaled problems, and the initial implementation provides a potential way for the linkage learning genetic algorithm to incorporate prior linkage information when such knowledge exists.

LNCS 3102, p. 971 ff.

Full article in PDF

lncs@springer.de
© Springer-Verlag Berlin Heidelberg 2004