 |
    |
A Clustering Based Niching Method for Evolutionary Algorithms
Felix Streichert1, Gunnar Stein2, Holger Ulmer1, and Andreas Zell1
1Center for Bioinformatics Tübingen (ZBIT),
University of Tübingen,
Sand 1,
72074 Tübingen, Germany,
streiche@informatik.uni-tuebingen.de,
http://www-ra.informatik.uni-tuebingen.de
2Institute of Formal Methods in Computer Science (FMI),
University of Stuttgart,
Breitwiesenstr. 20/22,
D-70565 Stuttgart, Germany,
http://www.informatik.uni-stuttgart.de/ifi/fk/index_e.html
Abstract. We propose the Clustering Based Niching (CBN) method for Evolutionary Algorithms (EA) to identify multiple global and local optima in a multimodal search space. The basic idea is to apply the biological concept of species in separate ecological niches to EA to preserve diversity. We model species using a multipopulation approach, one population for each species. To identify species in a EA population we apply a clustering algorithm based on the most suitable individual geno-/phenotype representation. One of our goals is to make the niching method as independent of the underlying EA method as possible in such a way that it can be applied to multiple EA methods and that the impact of the niching method on the EA mechanism is as small as possible.
LNCS 2723, p. 644 f.
lncs@springer.de
© Springer-Verlag Berlin Heidelberg 2003