Diversity control to improve convergence rate in genetic algorithms

Abstract

The diversity of the population affects the convergence rate in genetic algorithms. The determination of the proper diversity is still a trial and error process. The objective of this work is to study a method to find suitable population diversity automatically for a given problem. The proposed method is based on a modified restricted mating. A strategy to use diversity control is suggested using multiple subpopulations. Three well-known test problems, which have different requirement for diversity, are used to evaluate the proposed method. © Springer-Verlag 2003.