An Analysis of Minimum Population Search on Large Scale Global Optimization

Abstract

Minimum Population Search is a recently developed metaheuristic specifically designed to optimize high dimensional multi-modal functions. In this paper, Minimum Population Search is evaluated on the test functions provided for the 2013 LSGO competition in the IEEE Congress of Evolutionary Computation (CEC 2013). Furthermore, an analysis using some recent categorizations for exploitation and exploration, and especially the effects of selection, sheds light on how MPS performs better in high dimensions than popular metaheuristics such as Differential Evolution and Particle Swarm Optimization.