A Population Set-based Global Optimization Procedure Characterized by a Births Control Strategy

C. Bruni, C. Ferrone, and M. Lucchetti (Italy)

Keywords

:Genetic Algorithms, Evolutionary Algorithms, Controlled Random Search, Global optimization.

Abstract

A global optimization procedure which implements a population-set based approach is presented. It is based on a dynamical model of population, which includes control actions on the number of births and deaths, with the aim of intensifying the search in the most promising regions of the admissible set. The generation of newborns uniformly explores the compartments constituting the admissible set by exploiting the multidimensional Weyl theorem. The algorithm has been tested against several 2,3,4 dimensional classical test functions with promising and competitive results.

Important Links:



Go Back