基于Q学习的适应性进化规划算法  

An Adaptive Evolutionary Programming Algorithm Based on Q Learning

作　　者：张化祥[1] 陆晶[2] 

机构地区：[1]山东师范大学计算机系,济南250014 [2]山东财政学院计算机系,济南250014

出　　处：《自动化学报》2008年第7期819-822,共4页Acta Automatica Sinica

基　　金：国家自然科学基金(90612003);山东省中青年科学家科研奖励基金(2006BS01020);山东省自然科学基金(Y2007G16)资助~~

摘　　要：进化规划中,个体选择变异策略特别重要.适应性变异策略因在进化过程中动态选择个体变异策略,能够取得较好的性能.传统适应性变异策略都依据个体一步进化效果考察个体适应性,没有从多步进化效果上对变异策略进行评价.本文提出一种新的基于Q学习的适应性进化规划算法QEP(Q learning based evolutionary programming),该算法将变异策略看成行动,考察个体多步进化效果,并通过计算Q函数值,学习个体最优变异策略.实验表明,QEP能够获得好的性能.Selection of mutation strategies plays an important role in evolutionary programming, and adaptively selecting a mutation strategy in each evolutionary step can achieve good performance. A mutation strategy is evaluated and selected only based on the one-step performance of mutation operators in classical adaptive evolutionary programming, and the performance of mutation operators in the delayed mutation steps is ignored. This paper proposes a novel adaptive mutation strategy based on Q learning-- QEP （Q learning based evolutionary program- ming）. In this algorithm, several candidate mutation operators are used and each is considered as an action. The evolutionary performance of delayed mutation steps is considered in calculating the Q values for each mutation operator and the mutation operator that maximizes the learned Q values is the optimal one. Experimental results show that the proposed mutation strategy achieves better performance than the existing algorithms.

关 键 词：进化规划 变异策略 Q学习 收益 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]