计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (11A): 250100055-7.doi: 10.11896/jsjkx.250100055

• 人工智能 • 上一篇    下一篇

一种新的基于生长神经气体网络的多模态多目标优化算法

宣贺君1,2, 寇丽博1, 刘如意3   

  1. 1 信阳师范大学计算机与信息技术学院 河南 信阳 464000
    2 河南省教育大数据分析与应用重点实验室 河南 信阳 464000
    3 西安电子科技大学计算机科学与技术学院 西安 710071
  • 出版日期:2025-11-15 发布日期:2025-11-10
  • 通讯作者: 宣贺君(xuanhejun0896@xynu.edu.cn)
  • 基金资助:
    国家自然科学基金青年基金(62202366);河南省重点研发专项(241111212200);河南省科技研发计划联合基金项目(20240012);河南省教育课程改革研究项目(2025-JSJYYB-029)

Novel Multi-modal Multi-objective Algorithm Based on Growing Neural Gas Network

XUAN Hejun1,2, KOU Libo1, LIU Ruyi3   

  1. 1 School of Computer and Information Technology,Xinyang Normal University,Xinyang,Henan 464000,China
    2 Henan Key Laboratoray of Education Big Data Analysis and Application,Xinyang,Henan 464000,China
    3 School of Computer Science and Technology,Xidian University,Xi’an 710071,China
  • Online:2025-11-15 Published:2025-11-10
  • Supported by:
    National Natural Science Foundation of China(62202366),Henan Province Key Research and Development Project(241111212200),Henan Joint Fund for Science and Technology Research(20240012) and Teacher Education Curriculum Reform Research Project of Henan Province(2025-JSJYYB-029).

PDF (PC)

14

摘要： 多模态多目标优化是同一个Pareto前沿具有多个Pareto解集的复杂多目标优化问题,已成为多目标优化领域中的重要研究方向。已有的算法能够较好地解决该问题,但在解的多样性、收敛性及处理目标冲突方面表现出一定的局限性,如难以有效覆盖所有解集或在优化过程中出现收敛过早的现象。为解决这些问题,提出了一种新的基于生长神经气体网络(Growing Neural Gas,GNG)的环境选择策略的多模态多目标优化算法。该方法通过引入自适应拓扑结构,动态调整种群分布,同时利用加权的欧氏距离计算拥挤度以进行环境选择,提高种群的多样性和均匀性。此外,引入知识转移机制增强算法搜索能力,进一步提高解的多样性和收敛性。为验证算法的有效性,在HYL和MMF测试函数集上进行了实验。实验结果表明:所提算法在解的分布均匀性、Pareto前沿的收敛性及目标空间的覆盖性等方面的表现均优于5种对比算法。

关键词: 多模态, 多目标, 神经网络, 知识转移, 环境选择

Abstract: Multi-modal multi-objective optimization is a complex multi-objective optimization problem with multiple Pareto solutions on the same Pareto front.It has become an important research direction in the field of multi-objective optimization.Existing algorithms can solve this problem well,but they have certain limitations in terms of solution diversity,convergence and handling of target conflicts,such as difficulty in effectively covering all solution sets or premature convergence during the optimization process.To solve these problems,a new multi-modal multi-objective optimization algorithm based on the environment selection strategy of the growing neural gas(GNG) network is proposed.This method introduces an adaptive topological structure to dynamically adjust the population distribution,and uses weighted Euclidean distance to calculate the crowding degree for environment selection,thereby improving the diversity and uniformity of the population.In addition,the knowledge transfer mechanism is introduced to enhance the algorithm’s search ability and further improve the diversity and convergence of solutions.To verify the effectiveness of the algorithm,experiments are carried out on the HYL and MMF test function sets.The experimental results show that the proposed algorithm performs better than the five comparison algorithms in terms of solution distribution uniformity,Pareto front convergence and target space coverage.

Key words: Multi-modality, Multi-objective, Neural network, Knowledge transfer, Environmental selection

中图分类号: 

  • TP301
[1]BI J X,LI S Q,LIU J L,et al.Research of UWB coplanar base station deployment in indoor typical scenarios[J].Science of Surveying and Mapping,2024,49(3):19-26.
[2]ZHANG J N,HAN C C,CHEN J W,et al.A Method for Joint Edge Server Deployment and Service Placement[J].Computer Engineering,2024,50(10):266-280.
[3]LIU W C,WANG L P.Optimization scheme for low-powerUAV backscattering mobile edge computing network[J].Journal of XIAN University of Posts and Telecommunication,2024,29(5):38-46.
[4]XUAN H J,KOU L B,DING Y.et al.Multi-modal multi-objective optimization algorithm based on L1/2-norm crowding measurement.Journal of Xinyang Normal University[J/OL].[2024-12-18].http://kns.cnki.net/kcms/detail/41.1107.N.1107.002.html.
[5]HU K Q,MA W B,DAI C F,et al.Federated Learning Evolutionary Multi-objective Optimization Algorithm Based on Improved NSGA-III[J/OL].[2024-12-18].http://kns.cnki.net/kcms/detail/50.1075.TP.20240912.0844.002.html.
[6]JIANG R,FAN S W,WANG X M,et al.Clustering algorithm based on improved SOM model[J/OL].http://kns.cnki.net/kcms/detail/50.1075.TP.20241101.1447.032.html.
[7]LIANG J,QIAO K J,YUEC T,et al.A Clustering-Based Diffe-rential Evolution Algorithm for Solving Multi-Modal Multi-Objective Optimization Problems[J].Swarm and Evolutionary Computation,2021,60:100788.
[8]ZHANG K,SHEN C N,HE J J,et al.Knee Based Multi-Modal Multi-Objective Evolutionary Algorithm for Decision Making[J].Information Sciences,2021,544:39-55.
[9]LIU Y P,GARY Y.A Multi-modal Multi-objective Evolutiona-ry Algorithm Using Two-Archive and Recombination Strategies[J].IEEE Transactions on Evolutionary Computation,2019,23(4):660-673.
[10]ZHAO H,TANG L,LI J R,et al.Strengthening Evolution-BasedDifferential Evolution with Prediction Strategy for Multi-Modal Optimization and Its Application in Multi-Robot Task Allocation[J].Applied Soft Computing,2023,139:110218.
[11]LI Z H,SHI L,YUE C T,et al.Differential Evolution based on Reinforcement Learning with Fitness Ranking for Solving Multi-Modal Multi-Objective Problems[J].Swarm and Evolutionary Computation,2019,49:234-244.
[12]YUE C T,QU B Y,LIANG J.A Multi-objective Particle Swarm Optimizer Using Ring Topology for Solving Multi-modal Multi-objective Problems[J].IEEE Transactions on Evolutionary Computation,2018,22(5):805-817.
[13]QIAO K J,LIANG J,YU K,et al.Evolutionary constrainedmulti-objective optimization:Scalable high-dimensional constraint benchmarks and algorithm[J].IEEE Transactions on Evolutionary Computation,2024,28(5):965-979.
[14]LIN H,LIANG J,YUE C T,et al.A Niching-Based Reproduction and Preselection-Based Multi-objective Differential Evolution for Multimodal Multi-objective Optimization[C]//2024 IEEE Congress on Evolutionary Computation(CEC).IEEE,2024:1-8.
[15]LIANG J,SUI X,YUE C T,et al.Multimodal multi-objectivedifferential evolution algorithm based on enhanced decision space search[J].Swarm and Evolutionary Computation,2024,90:101682.
[16]CHEN P,LI Z,QIAO K J,et al.An archive-assisted multi-modal multi-objective evolutionary algorithm[J].Swarm and Evolutionary Computation,2024,91:101738.
[17]YUE C T,YE W H,ZHANG Y J,et al.Multimodal Multi-objective Optimization Algorithm Based on Local Center Clustering[J/OL].[2024-12-18].http://kns.cnki.net/kcms/detail/50.1075.tp.20241028.1146.021.html.
[18]LI H D,HU H,JIANG Q Q.Multimodal Multi-objective Optimization Based on Parallel Zoning Search and Its Application[J].Computer Science,2022,49(5):212-220.
[19]MING F,GONG W Y,JIN Y C.Growing Neural Gas Network-based surrogate-assisted Pareto set learning for multimodal multi-objective optimization[J].Swarm and Evolutionary Computation,2024,87:101541.
[20]ISHIBUCHI H,PENG Y,PANG L M.Multi-modal multi-ob-jective test problems with an infinite number of equivalent pareto sets[C]//2022 IEEE Congress on Evolutionary Computation(CEC).2022:1-8.
[21]ZHOU A M,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 Trans.Evol.Computter,2009,13(5):1167-1189.
[22]LINAG J,YUE C T,QU B Y,Multimodal multi-objective optimization:A preliminary study[C]//2016 IEEE Congress on volutionary Computation(CEC).2016:2454-2461.
[23]LIU Y,YEN G G,GONG D.A multimodal multiobjective evolutionary algorithm using two-archive and recombination strategies.IEEE Trans.Evol.Computter,2019(23):660-674.
[24]LI W H,MING M J,ZHANG T,et al.Multimodal multi-objective evolutionary algorithm considering global and local pareto fronts[J].Acta Automatica.Sinica,2023,49(1):148-160.
[25]ALCALÁ-FDEZ J,SANCHEZ L,GARCIA S,et al.Keel:a software tool to assess evolutionary algorithms for data mining problems[J].Soft Computer,2009(13):307-318.
[26]ZENG F,YANG T,YAO S.From Point Cloud to Triangular Mesh by Growing Neural Gas[J].Journal of Software,2013,24(3):651-662.
[27]SHENG X J,WU Y M,LI S B.Time series data predictionmethod for aluminum electrolysis process based on GNG-ANFIS[J].Computer Integrated Manufacturing Systems,2023,29(10):3239-3248.
[28]XUE M,WANG P,TONG X R.Enhanced Growing Neural GasBased Many-Objective Evolutionary Algorithm[J].Journal of Data Acquisition and Processing,2024,39(3):634-648.
[29]LIU Y,ZHANG L,ZENG X,et al.Evolutionary multimodalmultiobjective optimization guided by growing neural gas[J].Swarm and Evolutionary Computation,2024,86:101500.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路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