Fuzzy Control of Population Size in Evolutionary Algorithms
Oscar Castillo, Patricia Melín, Oscar Montiel · ASME Press eBooks · 2006
We are presenting a method for improving the performance of evolutionary algorithms (EAs). It consists of an intelligent method based in human expertise to establish a fuzzy system with the purpose of making more efficient the exploration and exploitation of the landscape by increasing or diminishing the amount of individuals through generations. This is done depending on the values of two measures, the variance value, and the amount of generations in which the evolution has not had a significant increase in the fitness value of the best individual.