MINIMUM POPULATION SEARCH – A SCALABLE METAHEURISTIC FOR MULTI-MODAL PROBLEMS

Abstract

Minimum Population Search is a new metaheuristic specifically designed for optimization of multi-modal problems. Its core idea is to guarantee full coverage of the search space with the smallest possible population. A small population increases the chances of convergence and the efficient use of function evaluations, but it can also induce the risk of premature convergence. To control convergence and provide diversification, thresheld convergence is used as a main component of this new metaheuristic. Computational results show that Minimum Population Search performs competitively against Particle Swarm Optimization, Differential Evolution, and Univariate Marginal Distribution Algorithm on a broad range of multi-modal problems