211 実数値GAのためのドロネー三角形分割を用いた子個体生成手法

抄録

To design crossover operators with high search ability in real-coded Genetic Algorithms, it will be efficient to introduce a mechanism, which can concentrate offspring in regions with a satisfactory evaluation value, to a conventional crossover operator based on the functional specialization hypothesis, which is an efficient guideline for design of crossover operators. Here, we propose a new offspring generation method using Delaunay triangulation. The proposed method is based on Simplex Crossover, which is a typical crossover operator based on the functional specialization hypothesis. In addition, the mechanism to concentrate offspring in regions with a satisfactory evaluation value is introduced by utilizing the Delaunay triangulation. Through numerical examples, the proposed method was shown to be capable of deriving the optimum with a smaller population size and lower number of evaluations than Simplex Crossover.