A New Model of Parallel Distributed Genetic Algorithms for Cluster Systems: Dual Individual DGAs
A New Model of Parallel Distributed Genetic Algorithms for Cluster Systems: Dual Individual DGAs
复制标题
集群系统并行分布式遗传算法的新模型：双个体DGA
DOI：
10.1007/3-540-39999-2_36
复制
发表时间：
2000
期刊：
影响因子：
--
通讯作者：
Y. Tanimura
中科院分区：
文献类型：
--
作者：
T. Hiroyasu;M. Miki;Masahiro Hamasaki;Y. Tanimura
提出了一种新的并行分布式遗传算法模型——双个体分布式遗传算法（DuDGA）。该算法使用户不必设置一些参数，因为每个分布式遗传算法（DGA）的孤岛只有两个个体。DuDGA可以自动确定交叉率、迁移率和岛屿数量。此外，与简单的GA和DGA方法相比，DuDGA可以用更少的分析找到更好的解。通过四个典型的数值测试函数，讨论了DuDGA方法的能力和有效性。
A new model of parallel distributed genetic algorithm, Dual Individual Distributed Genetic Algorithm (DuDGA), is proposed. This algorithm frees the user from having to set some parameters because each island of Distributed Genetic Algorithm (DGA) has only two individuals. DuDGA can automatically determine crossover rate, migration rate, and island number. Moreover, compared to simple GA and DGA methods, DuDGA can find better solutions with fewer analyses. Capability and effectiveness of the DuDGA method are discussed using four typical numerical test functions.