《计算机应用研究》|Application Research of Computers

熵值理论在多目标演化中的应用研究

Application and research of entropy theory in multi-objective evolution

免费全文下载 (已被下载 次)  
获取PDF全文
作者 张雨真,戴光明,彭雷,王茂才
机构 中国地质大学武汉 计算机学院,武汉 430074
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2013)12-3652-05
DOI 10.3969/j.issn.1001-3695.2013.12.034
摘要 为了克服传统多目标演化算法在进化后期遗传操作可能失效使算法性能降低以及基于概率建模的多目标算法在演化初期由于种群分布尚未呈现一定规律, 采样产生的新个体的搜索方向同目标方向存在差异, 提出一种基于熵值的多目标演化算法(entropy based multi-objective evolutionary algorithm, EB-MOEA)。算法利用种群进化过程中, 个体分布存在从无序到有序的现象, 设计了一种基于熵值理论的种群分布计算方法, 并将其作为种群从无序到有序过渡的判定准则, 指导遗传操作和概率建模操作切换的时机。新算法采用ZDT、DTLZ系列测试集进行实验, 通过与NSGA-Ⅱ以及RM-MEDA算法的实验对比, 证明了新判断准则的有效性, EB-MOEA具有更好的寻优性能。
关键词 多目标演化;熵;判定准则;基于熵值的多目标演化算法
基金项目 “十二五”民用航天专业技术预先研究项目,国家自然科学基金资助项目(61103144,60873107)
中国博士后科学基金资助项目(2011M501260,2012T50681,2012M511301)
湖北省自然科学基金资助项目(2010CDB04104,2011CDB348)
中央高校基本科研业务费专项资金资助项目(CUG120114)
本文URL http://www.arocmag.com/article/01-2013-12-034.html
英文标题 Application and research of entropy theory in multi-objective evolution
作者英文名 ZHANG Yu-zhen, DAI Guang-ming, PENG Lei, WANG Mao-cai
机构英文名 School of Computer Science, China University of Geosciences, Wuhan 430074, China
英文摘要 In the later stage of multi-objective evolutionary algorithm, the traditional genetic operation may be invalid, thus the performance of the algorithm will be reduced, while at the early stage of probabilistic modeling, the search direction of the sampling new individuals may be differ from the exact direction for the lack of distribution rule of the population. The paper proposed an entropy-based multi-objective evolutionary algorithm(EB-MOEA), it used the phenomenon that the population went from disorder to order in the process of evolution. It designed a distribution calculation method and used as the criteria which could guide the switching time of genetic operation and probabilistic modeling operation. The new algorithm adopted ZDT, DTLZ test suits to conduct the comparison experiment with NSGA-Ⅱ and RM-MEDA, results show that the effectiveness of the new criterion and the proposed algorithm has better optimization performance.
英文关键词 multi-objective evolution; entropy; criteria; EB-MOEA
参考文献 查看稿件参考文献
  [1] COELLOCOELLO C A. Evolutionary multi-objective optimization:a historical view of the field[J] . Computational Intelligence Magazine, 2006, 1(1):28-36.
[2] GONG Mao-guo, JIAO Li-cheng, YANG Dong-dong, et al. Research on evolutionary multi-objective optimization algorithms[J] . Journal of Software, 2009, 20(2):271-289.
[3] 李艳芝. 基于模型多目标原理的星座优化算法设计[D] . 武汉:中国地质大学(武汉), 2010.
[4] OKABE T, JIN Yao-chun, SENDHOFF B, et al. Voronoi-based estimation of distribution algorithm for multi-objective optimizaton[C] //Proc of IEEE Congress on Evolutionary Computation. 2004:1594-1601.
[5] BOSMAN P A N, THIERENS D. The naive MIDEA:a baseline multi-objective EA[C] //Proc of the 3rd International Conference on Evolutionary Multi-Criterion Optimization EMO. 2005:428-442.
[6] DONG Wei-shan, YAO Xin. Unified eigen analysis on multivariate Gaussian based estimation of distribution algorithms[J] . Information Sciences, 2008, 178(15):3000-3023.
[7] LAUMANNS M, OCENASEK J. Bayesian optimization algorithms for multi-objective optimization[C] //PPSN VII Proc of the 7th International Conference on Parallel Problem Solving from Nature. 2002:298-307.
[8] PELIKAN M, SASTRY K, GOLDBERG D. Multiobjective hBOA clustering, and scalability[C] //Proc of Conference on Genetic and Evolutionary Computation. 2005:663-670.
[9] ZHOU Ai-min, ZHANG Qing-fu, JIN Yao-chu, et al. A model-based evolutionary algorithm for bi-objective optimization[C] //Proc of Congress on Evolutionary Computation. 2005:2568-2575.
[10] ZHOU Ai-min, JIN Yao-chu, ZHANG Qing-fu, et al. Combining model-based and genetics-based offspring generation for multi-objective optimization:using a convergence criterion[C] //Proc of Congress on Evolutionary Computation. 2006:3234-3241.
[11] ZHANG Qing-fu, ZHOU Ai-min, JIN Yao-chu. RM-MEDA:a regularity modelbased multiobjective estimation of distribution algorithm[J] . IEEE Trans on Evolutionary Computation, 2008, 12(1):41-63.
[12] ZHOU Ai-min, ZHANG Qing-fu, JIN Yao-chu. Approximating the set of Pareto optimal solutions in both the decision and objective spaces by an estimation of distribution algorithm[J] . IEEE Trans on Evolutionary Computation, 2009, 13(5):1167-1189.
[13] ZHOU Ai-ming, QU Bo-yang, LI Hui, et al. Multiobjective evolutionary algorithm:a survey of the state of the art[J] . Swarm and Evolutionary Computation, 2011, 1(1):32-49.
[14] MO Li, DAI Guang-ming, ZHU Jian-kai. The RM-MEDA based on elitist strategy[C] //Proc of the 5th International Conference on Advances in Computation and Intelligence ISICA. 2010:229-239.
[15] 张冬梅, 龚小胜, 戴光明. 基于多重分形主曲线模型多目标演化算法研究[J] . 计算机研究与发展, 2011, 48(9):1729-1739.
[16] 郑金华. 多目标进化算法及其应用[M] . 北京:科学出版社, 2007.
[17] OCENASEK J. Entropy-based convergence measurement in discrete estimation of distribution algorithms[M] //Towards a New Evolutionary Computation:Advances in the Estimation of Distribution Algorithms. 2006:39-50.
[18] WANG Yao-nan, WU Liang-hong, YUAN Xiao-fang. Multiobjective self-adaptive difference evolution with elitist archive and crowding entropy-based diversity measure[J] . Soft Computing, 2010, 14(3):193-209.
[19] QIN Yu-fang, JI Jun-zhong, LIU Chun-nian. An entropy-based multiobjetive evolutionary algorithm with an enhanced elite mechanism[J] . Applied Computational Intelligence and Soft Computing, 2012, 2012:Article No. 17.
[20] WANG Lin-lin, CHEN Yun-fang. Diversity based on entropy:a novel evaluation criterion in multi-objective optimization algorithm[J] . Intelligent Systems and Appications, 2012, 4(10):113-124.
[21] ZITZLER E, DEB K, THIELE L. Comparison of multiobjective evolutionary algorithms:empiriCal results[J] . Evolutionary Computation, 2000, 8(2):173-195.
[22] DEB K, PRATAP A, MEYARIVAN T. A fast and elitist multiobjective genetic algorithm:NSGA-Ⅱ[J] . Evolutionary Computation, 2002, 6(2):182-197.
收稿日期
修回日期
页码 3652-3656
中图分类号 TP391
文献标志码 A