机构地区: 深圳大学信息工程学院
出 处: 《深圳大学学报(理工版)》 2004年第4期306-309,共4页
摘 要: 对遗传操作的选择策略进行了理论和实验研究.以可编程树型结构函数为进化硬件的实验目标模型,对精英保留、锦标赛选择和赌轮选择策略进行了分析与实验,探讨可节省硬件资源消耗,且可加速搜索的适于硬件化的选择策略. The evolvable hardware in general and the selection strategies of genetic algorithm in particular are studied. Theoretical and experimental studies are carried out on three different select strategies: elitist selection, tournament selection and roulette wheel selection, with tree structure function as object to be evolved. Performance of different select strategies is compared and some general guidelines on determining selection strategy for evolutionary algorithm in research on evolvable hardware are given.
领 域: [自动化与计算机技术] [自动化与计算机技术]