Abstract
This paper presents methods to retain diversity of the allele distribution in the search of Genetic Algorithms. The purpose of this study is to refine the convergence behaviors of Genetic Algorithms for use in multi-variable optimization. We first evaluate“rareness”of a chromosome using the distribution of alleles in a population. Then the fitness of a chromosome is adjusted with the rareness so that rare chromosomes will be likely to survive. In addition, “mutation width”is introduced to tune the effect of the mutation which can generate rare chromosomes while conserving effective convergence.
