超一様分布列を用いることによる遺伝的アルゴリズムの性能改善

抄録

The random number generator is one of the important components of evolutionary algorithms. Therefore, when we try to solve function optimization problems using the evolutionary algorithms, we must carefully choose a good pseudo-random number generator. In the evolutionary algorithms, the pseudo-random number generator is often used for creating uniformly distributed individuals. In this study, as the low-discrepancy sequences allow us to create individuals more uniformly than the random number sequences, we apply the low-discrepancy sequence generator, instead of the pseudo-random number generator, to the evolutionary algorithms. Since it was difficult for some evolutionary algorithms, such as binary-coded genetic algorithms, to utilize the uniformity of the sequences, the low-discrepancy sequence generator was applied to real-coded genetic algorithms. The numerical experiments show that the low-discrepancy sequence generator improves the search performances of the real-coded genetic algorithms.