A. Mayer and H.A. Mayer (Austria)
Neuroevolution, multi–chromosomal representation, ANN classification, ANN prediction
Contemplating the development of the field of evolu tionary computation (EC), where most basic concepts are borrowed from nature, it is remarkable that multi– chromosomal representations present in all complex or ganisms have rarely been studied in the artificial domain. Evidently, the addition of such an additional layer of genetic code must prove to possess certain benefits be fore it is introduced. Thus, we present experiments with multi–chromosomal evolution of artificial neural networks (ANNs) on three benchmark problems with the goal to in vestigate potential advantages of genotypes with multiple chromosomes. Besides the technical benefit of using differ ent encodings, genetic operators, and parameters for spe cific chromosomes, we hypothesize that the generalization capabilities of evolved networks (or other structures) in crease, when their genotype is of multi–chromosomal or ganization.
Important Links:
Go Back
