Edit distance based crossover operator in gene expression programming

Abstract

The population diversity greatly affects the evolutionary efficiency and solution quality of gene expression programming algorithm. Population diversity should be preserved by keeping certain distance between individuals in the population. Edit distance can describe the similarity of individuals well. Crossover is a way to create and maintain the distance of the individuals. In this paper, we propose two edit distance based crossover operators. Experimental results show that the proposed farthest edit distance based crossover operator is able to preserve the diversity of population and solve the optimization problem more efficiently.