Discussion of the Three Problems in Genetic Algorithm with Chromosome Differentiation

摘要

Bandyopadhyay and Pal proposed an improved genetic search strategy, GACD (Genetic Algorithm with Chromosome Differentiation), involving partitioning the chromosomes into two classes, and defining a restricted form of the crossover operator between the two classes. The GACD can be applied to many multi-dimensional pattern recognition problems. However, their GACD suffered from three problems, i.e. ȁC;sampling-without-replacementȁD;, ȁC;mating-with-best-chromosomeȁD; and ȁC;the-unreachable-resultȁD;.