®®®® SIIA Público

Título del libro: Proceedings Of The 12th Annual Genetic And Evolutionary Computation Conference, Gecco '10
Título del capítulo: Why recombination should be adaptive

Autores UNAM:
CHRISTOPHER RHODES STEPHENS;
Autores externos:

Idioma:
Inglés
Año de publicación:
2010
Palabras clave:

Adaptation; Adaptive crossovers; Building blockes; Crossover operator; Crossover points; Genetic operators; Random selection; Recombination; Test functions; Genetic algorithms; Mathematical operators


Resumen:

Crossover is the most complicated of the standard genetic operators as well as one of the main operators in genetic algorithms. The role of the crossover operator has not been satisfactorily explained. It has not been easy to show if it is essential for the construction and exploitation of building blocks during evolution. In this paper we introduce an adaptive crossover operator for two test functions. The results show that there is a clear increase in the rate of evolution when information on the state of the population is used to select the crossover point, compared to random selection of the crossover point.


Entidades citadas de la UNAM: