Computer Engineering and Applications ›› 2015, Vol. 51 ›› Issue (11): 12-17.

Previous Articles     Next Articles

Research on effects of misleading information to performance of EAs

LI Kun1, LI Ming1,2, CHEN Hao2   

  1. 1.College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
    2.School of Information Engineering, Nanchang Hangkong University, Nanchang 330063, China
  • Online:2015-06-01 Published:2015-06-12


李  坤1,黎  明1,2,陈  昊2   

  1. 1.南京航空航天大学 自动化学院,南京 210016
    2.南昌航空大学 信息工程学院,南昌 330063

Abstract: This paper studies effects of the misleading information in the fitness landscape to the performance of evolutionary algorithm. The normal description for the optimizing process of evolutionary algorithm is proposed on the basis of optimal contraction theorem. The gradient information related set is proposed to character the effects of different kinds of misleading information. Accordingly, the effects of misleading information are fallen into two categories:deception and multi-peaks, and their influences to the evolutionary algorithm are theoretically analyzed. The theoretical analyzing results are verified and complemented by testing some real-parameter function.

Key words: fitness landscape, optimal contraction theorem, particle swarm optimization, genetic algorithm

摘要: 以适应值曲面模型为基础分析误导信息对进化算法性能的影响。基于最优吸引子提出描述进化算法作用过程的一般方法,提出梯度信息关联集合的概念以描述误导信息的种类和作用。在此基础上,将误导信息的作用分为欺骗和多峰两类,并分别对其作用和影响进行理论分析。通过测试数值函数验证并补充理论分析的结果。

关键词: 适应值曲面, 最优吸引子理论, 粒子群算法, 遗传算法