计算机科学

计算机科学 ›› 2026, Vol. 53 ›› Issue (6A): 250500081-11.doi: 10.11896/jsjkx.250500081

• 人工智能 • 上一篇    下一篇

基于MOEA/D训练PF模型的多目标进化算法

李笠, 易佳丽, 李攸俊, 李广鹏   

  1. 桂林电子科技大学广西可信软件重点实验室 广西 桂林 541004
  • 出版日期:2026-06-16 发布日期:2026-06-12
  • 通讯作者: 李笠(lili@guet.edu.cn)
  • 基金资助:
    国家自然科学基金(62366009)

Multi-objective Evolutionary Method by Training Front Modeling Based on MOEA/D

LI Li, YI Jiali, LI Youjun, LI Guangpeng   

  1. Guangxi Key Laboratory of Trusted Software,Guilin University of Electronic Technology,Guilin,Guangxi 541004,China
  • Published:2026-06-16 Online:2026-06-12
  • About author:LI Li,born in 1986,Ph.D,associate professor,master's supervisor,is a member of CCF(No.39086M).His main research interest is multi-objective optimization methods and their applications.
  • Supported by:
    National Natural Science Foundation of China(62366009).

PDF (PC)

40

摘要： 基于分解的多目标进化算法被广泛用于解决多目标问题,然而当具有不同形状Pareto前沿(PF)的多目标问题选择了一个不合适的分解策略时,可能会产生不理想的结果。针对此情况,设计并采用一种基于MOEA/D的训练PF模型的多目标进化算法,即MOEA/D-ECM算法来解决分解策略对PF形状敏感的问题。该算法通过训练PF通用模型来预测PF曲率,然后根据预测的曲率来选择合适的分解策略,同时为了确保算法的分布性,在MOEA/D算法中加入小生境技术分布策略来选择交配亲本,以提高后代质量。为了评价该算法的性能,将该算法与几种多目标进化算法在具有凹状、凸状和线性PF的不同测试问题上进行了对比实验。实验结果表明,MOEA/D-ECM算法能够有效求解具有不同曲率的多目标优化问题,并且具有良好的性能和竞争力。

关键词: PF模型, 分解策略, 多目标优化问题, 适应度函数

Abstract: Multi-objective evolutionary algorithm based on decomposition(MOEA/D) is a widely employed optimization strategy in real-world applications.However,choosing a decomposition strategy of the MOEA/D that is not suitable for the curvature of Pareto Front(PF) can produce unsatisfactory results when dealing with multi-objective optimization problems.To address this issue,a multi-objective evolutionary algorithm based on MOEA/D by training PF models,named MOEA/D-ECM,is designed and adopted to solve the problem of the sensitivity of decomposition strategies to PF curvature.The algorithm trains a generic PF model to predict the curvature of the PF and then selects an appropriate decomposition strategy based on the predicted curvature.In addition,to ensure the diversity of the algorithm,a niche technique and distribution strategy is incorporated into the MOEA/D algorithm to select mating parents and improve the quality of the offspring.To evaluate the performance of this algorithm,several multi-objective evolutionary algorithms are compared on different test problems with concave,convex,and linear PF.The experimental results demonstrate that the MOEA/D-ECM algorithm can effectively solve multi-objective optimization problems for PF with different curvatures and has good performance and competitiveness.

Key words: PF model, Decomposition strategy, Multi-objective optimization problems, fitness function

中图分类号: 

  • TP181
[1] ZHOU A,QU B,LI H,et al.Multiobjective evolutionary algorithms:A survey of the state of the art[J].Swarm and Evolutionary Computation,2011,1(1):32-49.
[2] VACHHANI V L,DABHI V K,PRAJAPATI H B.Survey of multi objective evolutionary algorithms[C]//2015 International Conference on Circuits,Power and Computing Technologies(ICCPCT-2015).IEEE,2015:1-9.
[3] TRIVEDI A,SRINIVASAN D,SANYAL K,et al.A survey of multiobjective evolutionary algorithms based on decomposition[J].IEEE Transactions on Evolutionary Computation,2016,21(3):440-462.
[4] XU Q,XU Z,MA T.A survey of multiobjective evolutionary algorithms based on decomposition:variants,challenges and future directions[J].IEEE Access,2020,8:41588-41614.
[5] MA X,YU Y,LI X,et al.A survey of weight vector adjustment methods for decomposition-based multiobjective evolutionary algorithms[J].IEEE Transactions on Evolutionary Computation,2020,24(4):634-649.
[6] KE L,ZHANG Q,BATTITI R.MOEA/D-ACO:A Multiobjective Evolutionary Algorithm Using Decomposition and AntColony[J].IEEE Transactions on Cybernetics,2013,43(6):1845-1859.
[7] DONG Z,WANG X,TANG L.MOEA/D with a self-adaptive weight vector adjustment strategy based on chain segmentation[J].Information Sciences,2020,521:209-230.
[8] GENG H T,HAN W M,ZHOU S S,et al.MOEA/D Algorithm Based on New Neighborhood Updating Strategy [J].Computer Science,2019,46(5):191-197.
[9] DEB K,PRATAP A,AGARWAL S,et al.A fast and elitist multiobjective genetic algorithm:NSGA-II[J].IEEE Transactions on Evolutionary Computation,2002,6(2):182-197.
[10] WHILE L,HINGSTON P,BARONE L,et al.A faster algorithm for calculating hypervolume[J].IEEE Transactions on Evolutionary Computation,2006,10(1):29-38.
[11] ZHOU A,JIN Y,ZHANG Q,et al.Combining model-based and genetics-based offspring generation for multi-objective optimization using a convergence criterion[C]//2006 IEEE International Conference on Evolutionary Computation.IEEE,2006:892-899.
[12] TRAUTMANN H,WAGNER T,BROCKHOFF D.R2-EMOA:Focused multiobjective search using R2-indicator-based selection[C]//7th International Conference Learning and Intelligent Optimization(LION 7).Catania,Italy,Revised Selected Papers 7.Springer Berlin Heidelberg,2013:70-74.
[13] ZHANG Q,LI H.MOEA/D:A Multiobjective Evolutionary Algorithm Based on Decomposition[J].IEEE transactions on evolutionary computation,2007,11(6):712-731.
[14] GENG H T,DING Y Y,ZHOU L F,et al.Improved MOEA/D Algorithm Based on Self-adaptive Selection Strategy[J].Computer Science,2018,45(5):201-207.
[15] MORÉ J J.The Levenberg-Marquardt algorithm:implementa-tion and theory[C]//Numerical analysis:proceedings of the biennial Conference held at Dundee.Berlin,Heidelberg:Springer Berlin Heidelberg,2006:105-116.
[16] BOWMAN JR V J.On the relationship of the Tchebycheff normand the efficient frontier of multiple-criteria objectives[C]//Multiple Criteria Decision Making:Proceedings of a Conference Jouy-en-Josas.Berlin,Heidelberg:Springer Berlin Heidelberg,1976:76-86.
[17] JASZKIEWICZ A.On the performance of multiple-objective geneticlocal search on the 0/1 knapsack problem-a comparative experiment[J].IEEE Transactions on Evolutionary Computation,2002,6(4):402-412.
[18] LI L,YEN G G,SAHOO A,et al.On the estimation of pareto front and dimensional similarity in many-objective evolutionary algorithm[J].Information Sciences,2021,563:375-400.
[19] AMMAR A,SHERBINI A E,ASHOUR I,et al.Random data encryption algorithm(RDEA)[C]//Proceedings of the Twenty-Second National Radio Science Conference(NRSC 2005).IEEE,2005:359-366.
[20] XIANG Y,ZHOU Y,YANG X,et al.A many-objective evolutionary algorithm with Pareto-adaptive reference points[J].IEEE Transactions on Evolutionary Computation,2019,24(1):99-113.
[21] HERNANDEZ-DIAZ A G,SANTANA-QUINTERO L V,COELLO C C,et al.Pareto-adaptive epsilon-dominance[J].Evol Comput,2007,15(4):493-517.
[22] ZAPOTECAS MARTÍNEZ S,SOSA HERNÁNDEZ V A,AGUIRRE H,et al.Using a family of curves to approximate the Pareto front of a multi-objective optimization problem[C]//Parallel Problem Solving from Nature-PPSN XIII:13th International Conference,Ljubljana,Slovenia.Springer International Publishing,2014:682-691.
[23] ZHOU A,ZHANG Q,JIN Y.Approximating the set of Pareto-optimal solutions in both the decision and objective spaces by an estimation of distribution algorithm[J].IEEE Transactions on Evolutionary Computation,2009,13(5):1167-1189.
[24] TIAN Y,ZHANG X,CHENG R,et al.Guiding EvolutionaryMultiobjective Optimization With Generic Front Modeling[J].IEEE Transactions on Cybernetics,2020,50(3):1106-1119.
[25] WILAMOWSKI B M,YU H.Improved computation for Levenberg-Marquardt training[J].IEEE Trans Neural Netw,2010,21(6):930-937.
[26] GAVIN H P.The Levenberg-Marquardt algorithm for nonlinear least squares curve-fitting problems[D].Duke University Department of Civil and Environmental Engineering,2022.
[27] LUO B,ZHENG J,XIE J,et al.Dynamic crowding distance? A new diversity maintenance strategy for MOEAs[C]//2008 Fourth International Conference on Natural Computation.IEEE,2008,1:580-585.
[28] WEI L,ZHAO M.A niche hybrid genetic algorithm for global optimization of continuous multimodal functions[J].Applied Mathematics and Computation,2005,160(3):649-661.
[29] WANG R,ZHANG Q,ZHANG T.Decomposition-Based Algorithms Using Pareto Adaptive Scalarizing Methods[J].IEEE Transactions on Evolutionary Computation,2016,20(6):821-837.
[30] LIU H,GU F,ZHANG Q.Decomposition of a MultiobjectiveOptimization Problem Into a Number of Simple Multiobjective Subproblems[J].IEEE Transactions on Evolutionary Computation,2014,18(3):450-455.
[31] LIU Y,GONG D,SUN J,et al.A Many-Objective Evolutionary Algorithm Using A One-by-One Selection Strategy[J].IEEE Transactions on Cybernetics,2017,47(9):2689-2702.
[32] DEB K,THIELE L,LAUMANNS M,et al.Scalable test problems for evolutionary multiobjective optimization[M]//Evolutionary Multiobjective Optimization:Theoretical Advances and Applications.London:Springer London,2005:105-145.
[33] ZITZLER E,THIELE L,LAUMANNS M,et al.Performanceassessment of multiobjective optimizers:an analysis and review[J].IEEE Transactions on Evolutionary Computation,2003,7(2):117-132.
[34] LI L,LI G P,CHANG L,et al.Survey of Constrained Evolutionary Algorithms and Their Applications[J].Computer Science,2021,48(4):1-13.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市两江新区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发