Particle swarm optimization with dynamic search strategies based on landscape modality estimation

Abstract

The paper presents particle swarm optimization (PSO) with dynamic search strategies based on landscape modality estimation. In order to control search strategies, we introduce landscape modality estimation method using correlation coefficients between rankings of search points to PSO. This estimation method utilizes relationship between fitness and distance to a reference point to classify whether the landscape modality is uni-modal or multi-modal landscape. Our proposal method can switch the strategies properly according to landscape modality of an objective function. To confirm the search ability of the proposal method, we conducted experiments using standard benchmark functions. The experimental results show that the proposal method outperforms other PSO variants.