 |
    |
Towards a Generally Applicable Self-Adapting Hybridization of Evolutionary Algorithms
Wilfried Jakob1, Christian Blume2, and Georg Bretthauer1
1Forschungszentrum Karlsruhe, Institute for Applied Computer Science, Postfach 3640, 76021 Karlsruhe, Germany
wilfried.jakob@iai.fzk.de
georg.bretthauer@iai.fzk.de
2University of Applied Sciences, Cologne, Campus Gummersbach, Am Sandberg 1, 51643 Gummersbach, Germany
blume@gm.fh-koeln.de
Abstract. Practical applications of Evolutionary Algorithms (EA) frequently use some sort of hybridization by incorporating domain-specific knowledge, which turns the generally applicable EA into a problem-specific tool. To overcome this limitation, the new method of HyGLEAM was developed and tested extensively using eight test functions and three real-world applications. One basic kind of hybridization turned out to be superior and the number of evaluations was reduced by a factor of up to 100.
LNCS 3102, p. 790 f.
lncs@springer.de
© Springer-Verlag Berlin Heidelberg 2004