论文信息 - IMPROVING GENETIC ALGORITHMS PERFORMANCE BY HASHING FITNESS VALUES

IMPROVING GENETIC ALGORITHMS PERFORMANCE BY HASHING FITNESS VALUES

This paper presents a method for improving genetic algorithm (GA) performance. Typically, zero diversity in the population's fitness values signals the stopping point for a GA. As the population evolves, diversity diminishes, causing the same chromosomes to be frequently reevaluated. For real world problems, the computational effort spent on evaluating the fitness function far exceeds that of the genetic operators. By using a hash table to store the most recently evaluated chromosomes, significant performance improvements are realized. Several examples demonstrate the improvements.

Richard J. Povinelli | Xin Feng

[1] Richard J. Povinelli,et al. TEMPORAL PATTERN IDENTIFICATION OF TIME SERIES DATA USING PATTERN WAVELETS AND GENETIC ALGORITHMS , 1999 .

[2] David E. Goldberg,et al. Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[3] Udi Manber,et al. Introduction to algorithms - a creative approach , 1989 .

[4] John H. Holland,et al. Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .