Cultural algorithm based on adaptive cauchy mutated particle swarm optimizer for high-dimensional function optimization

This paper presents a novel cultural algorithm, in which an adaptive Cauchy mutated particle swarm optimizer (ACMPSO) is used as a population space; the Cauchy allows larger mutations and in this way producing more diversified individuals and covering more major space. The knowledge sources contained in the belief space are specifically designed according to the ACMPSO evolution features. Different Gaussian mutated knowledge sources are used to influence the variation operator of ACMPSO; Gaussian mutation is accomplished in accurate search of its nearest space for proximity exploration, and it performs better in small neighborhood. The simulation results of benchmark test functions show that the proposed algorithm has good optimization quality and searching efficiency, especially it is a promising way for complex functions optimization with high dimensions.