摘要
提出了一种理想化的模拟仿生搜索算法——扰动算法 ,以此方法为基础 ,分析了遗传算法的搜索过程和效率问题 ,阐明了遗传算法作为一种次优算法的有效性 .相对于遗传算法的生物解释 ,本文给出了相应的物理解释 .同时 ,本文为遗传算法、进化策略和模拟退火算法找到了一种统一的物理解释 ,揭示了这些重要的仿生类算法实质上的相似性 .
An ideal simulated biological search algorithm, so called perturbation algorithm, is presented in this paper. Based on the proposed algorithm, the search process and the efficiency of the genetic algorithm are analyzed and the effectiveness of the genetic algorithm as a suboptimal algorithm is demonstrated. According to the biological interpretation of the genetic algorithm, the corresponding physical interpretation is given. A kind of unitary physical interpretation is found for the genetic algorithm, the evolutionary strategies and the simulated annealing algorithm. The substantive similarities of the important simulated biological algorithms are revealed.
出处
《数学的实践与认识》
CSCD
北大核心
2004年第2期96-103,共8页
Mathematics in Practice and Theory
基金
教育部科学技术研究重点项目 ( 0 2 0 90 )资助
