一种基于边缘分布估计的多目标优化算法

李斌; 钟润添; 肖金超; 庄镇泉

doi:10.3724/SP.J.1146.2006.00638

一种基于边缘分布估计的多目标优化算法

doi: 10.3724/SP.J.1146.2006.00638

基金项目:

国家自然科学基金(60401015，60572012)和安徽省自然科学基金(050420201)资助课题

计量
- 文章访问数: 2978
- HTML全文浏览量: 81
- PDF下载量: 1275
- 被引次数: 0
出版历程
- 收稿日期: 2006-05-15
- 修回日期: 2007-01-30
- 刊出日期: 2007-11-19

A Multi-Objective Optimization Algorithm Based on Marginal Distribution Estimation

摘要

摘要: 该文提出了一种基于边缘分布估计的多目标优化算法，通过在每一进化代中估计较优个体的边缘概率分布来引导算法对Pareto最优解的搜索。通过与基于拥挤机制的多样性保持技术、基于非支配排序的联赛选择、精英保留等技术的有机结合，使得算法在具有良好收敛性能的同时，具有很好的维持群体多样性的能力。通过一组典型测试函数实验对该算法的性能进行了分析，并与NSGA-II、SPEA、PAES等知名多目标优化算法进行了比较，结果表明该文算法收敛速度较快，且得到的非支配解集分布均匀，适合于复杂多目标优化问题的求解。
- 多目标优化;边缘分布估计;非支配解;多样性
Abstract: A new multi-objective optimization algorithm based on marginal distribution estimation is proposed, in which marginal probability distribution of the selected better individuals is estimated and is used to guide the search of Pareto optimal solutions of the multi-objective optimization problems. Combined with non-dominant ranking, diversity preserving technique based on crowding mechanism, tournament selection based on non-dominant ranking, and elitist strategy, the algorithm achieves a good balance between convergence and diversity. A set of typical test functions are used to evaluate the performance of the proposed algorithm, and comparison is made between some well-known multi-objective optimization algorithms, i.e. NSGA-II, SPEA, PAES. The experimental results show that the proposed algorithm can achieve a good balance between convergence and diversity, and is suited to complex multi-objective problems.

HTML全文

参考文献(1)

Fonseca C M and Fleming P J. Genetic algorithms for multiobjective optimization: formulation , discussion and generation[C]. Proceedings of the 5th International Conference on Genetic Algorithms, San Mateo, California, 1993: 416-423.Horn J and Nafpliotis N. Multiobjective optimization using the niched Pareto genetic algorithm[R]. USA: IlliGAL Report 93005, 1993.[2]Srinivas N and Deb K. Multiobjective optimization using nondominated sorting in genetic algorithms[J].Evolutionary Computation.1994, 2 (3):221-248[3]Zitzler E, Deb K, and Thiele L.. Comparison of multiobjective evolutionary algorithms: Empirical results[J].Evolutionary Computation.2000, 8 (2):173-195[4]Deb K, Pratap A, and Agarwal T. A Fast and Elitist Multiobjective Genetic Algorithm: NSGA-II[J].IEEE Transactions on Evolutionary Computation.2002, 6(2):182-197[5]Pelikan M, Goldberg D E, and Lobo F. A survey of optimization by building and using probabilistic models[R]. IlliGAL Tech. Rep. 99018, 1999.[6]Mhlenbein H and Paa G. From recombination of genes to the estimation of distributions I. Binary parameters[C]. Parallel Problem Solving from Nature - PPSN IV, 1996: 178-187.[7]Pelikan M, Goldberg D E, and Cant-Paz E. BOA: The Bayesian optimization algorithmp[C]. In Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 1999), 13-17 July 1999, Orlando, Florida, USA, 1999, Vol I: 525-532.[8]Larraaga P.[J].Etxeberria R, Lozano J A, and Pea J M. Optimization by Learning and Simulation of Bayesian and Gaussian Networks[R]. Dept. Comput. Sci. Artific. Intell., Univ. Basque Country, Tech. Rep. EHUKZAA-IK-4/9.1999,:-[9]Mhlenbein H, and Paa G. From recombination of genes to the estimation of distributions I. Binary parameters. In Proc. 5th Parallel Problem Solving from Nature(PPSN V), Amsterdam, The Netherlands, September 27-30, 1998, LNCS 1141: 178.

施引文献

资源附件(0)

访问统计