演化算法行为分析──基于全局随机搜索的理论刻划

演化算法行为分析──基于全局随机搜索的理论刻划 AN ANALYSIS ABOUT BEHAVIOR OF EVOLUTIONARY ALGORITHMS: A KIND OF THEORETICAL DESCRIPTION BASED ON GLOBAL RANDOM SEARCH METHODS 丁立新 Ding Lixin 1 first-author 康立山 Kang Lishan 1 陈毓屏 Chen Yupin 1 周少泉 Zhou Shaoquan 1 武汉大学软件工程国家重点实验室 武汉大学软件工程国家重点实验室 State Key Laboratory of Software Engineering, Wuhan University State Key Laboratory of Software Engineering, Wuhan University 根据候选解空间上抽样分布的构造和计算来描述演化算法的行为，抽样分布的迭代构造是利用基于代（ｇｅｎｅｒａｔｉｏｎ）方法的全局随机搜索思想来刻划，在这种框架下，比例选择看成是全局随机搜索算子，复合看成是开发候选解相似性的搜索过程．研究表明：适当地限制复合算子的搜索宽度，能够保证演化算法弱收敛到全局最优解． Evolutionary computation is a kind of adaptive non numerical computation method which is designed to simulate evolution of nature. In this paper, evolutionary algorithm behavior is described in terms of the construction and evolution of the sampling distributions over the space of candidate solutions. Iterative construction of the sampling distributions is based on the idea of the global random search of generational methods. Under this frame, propontional selection is characterized as a gobal search operator, and recombination is characerized as the search process that exploits similarities. It is shown that by properly constraining the search breadth of recombination operators, weak convergence of evolutionary algorithms to a global optimum can be ensured. 全局随机搜索 演化算法 弱收敛 遗传算法 global random search evolutionary algorithms weak convergence genetic algorithms TP301.6 国家自然科学基金(61672102) 1998-01-01 2021-04-01 1