A comparative analysis of FSS with CMA-ES and S-PSO in ill-conditioned problems

Abstract

This paper presents a comparative analyzes between three search algorithms, named Fish School Search, Particle Swarm Optimization and Covariance Matrix Adaptation Evolution Strategy applied to ill-conditioned problems. We aim to demonstrate the effectiveness of the Fish School Search in the optimization processes when the objective function has ill-conditioned properties. We achieved good results for the Fish School Search and in some cases we obtained superior results when compared to the other algorithms. © 2012 Springer-Verlag.