An improved genetic algorithm for solving deceptive problems

Based on the research on deceptive problems of genetic algorithms (GAs), this paper proposes an improved GA to solve deceptive problems. This method, by defining the diversity measure of gene loci for elitist individuals, deduces some gene loci which may cause schema deception. Then, the algorithm performs the complementary operation for these gene loci of elitist individuals. At last, this paper selects the trap function to do some experiments, and the experimental results show that this method is effective to solve deceptive problems of GAs.