Computer Engineering and Applications ›› 2007, Vol. 43 ›› Issue (21): 80-83.

• 学术探讨 • Previous Articles     Next Articles

Amplitude-based coding hybrid quantum-inspired evolutionary algorithm

YANG Qing,DING Sheng-chao   

  1. School of Computer Science and Technology,South-Central University for Nationalities,Wuhan 430074,China
    Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100080,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-07-21 Published:2007-07-21
  • Contact: YANG Qing

基于概率幅编码的混合量子演化算法

杨 青,丁圣超   

  1. 中南民族大学 计算机科学学院,武汉 430074
    中国科学院 计算技术研究所,北京 100080
  • 通讯作者: 杨 青

Abstract: Coding the individual with amplitude,and applying the crossover operator of classical genetic algorithm to the evolutionary goals of quantum-inspired evolutionary algorithm,a hybrid quantum-inspired evolutionary algorithm is proposed.Combining the self-adaptive rotation,defining a self-adaptive mutation operator with respect to mutation degree,and exchanging the information of the evolutionary goals by crossover operator regularly,the novel algorithm avoids converging prematurely and yields high efficiency.Also the algorithm exceeds QEA and CGA when they solve the numerical optimization problems.The Need-in-a-Haystack problem can be figured out perfectly.Additionally,the algorithm’s computing speed is similar to the classical one.

Key words: hybrid algorithm, quantum-inspired evolutionary, crossover operator, mutation degree

摘要: 个体基于量子概率幅进行编码,并将经典遗传算法的杂交算子用于量子演化算法中演化目标的优化,提出了混合量子演化算法。算法中对量子旋转角自适应更新,并首次引入了突变度的概念定义了自适应的变异算子,对量子个体的演化目标定期实施杂交,有效地交换并利用了演化信息,避免了未成熟收敛,提高了算法效率。数值优化问题的实验结果表明该算法优于QEA和CGA,并能以极大概率成功地解决“大海捞针”问题,且计算效率高,优化速度与CGA相当。

关键词: 混合算法, 量子演化, 杂交算子, 突变度