Auto-Organização em Algoritmos Genéticos Aplicados a Problemas Difíceis

Renato Tinós

Genetic algorithms (GAs) where the individual with the worst fitness and its two next neighbors are replaced by random individuals in every generation are investigated in this paper. This simple approach can take the GA to a self-organization behavior, which can be useful in difficult problems to maintain the diversity of the solutions and, then, to allow the GA to escape from local optima. In order to avoid that the individuals with the best fitness take the new individuals to extinction, these ones are preserved in a subpopulation. The analysis of the experimental results suggests that the proposed GA presents a kind of self-organizing behavior, known as Self-Organized Criticality, which is present in several natural phenomena.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: