上官晨曦,时振涛

• 1:太原科技大学计算机科学与技术学院

摘要(Abstract)：

迁移学习是一种解决动态多目标优化问题的有效方法，但当源域和目标域差异性较大时，会产生负迁移，大大降低求解优化问题的效率。针对这种现象该算法提出一种基于源域选择策略的迁移学习方法。该方法首先根据历史环境的最优解集与新环境目标域的差异性对历史环境数据进行排序，选择一个差异性最小的历史环境数据作为迁移解；同时，对t时刻环境的最优解集进行交叉变异生成多样性解，将其与迁移解合进行非支配排序得到源域数据；然后将源域数据映射到嵌入空间，求出最优解作为新环境下的初始种群进行下一时刻迭代运算。这种方法考虑了多个历史环境知识的重用，可以加强种群全局搜索能力，有效抑制负迁移的产生，从而提高算法效率。通过实验，结果证明本文提出的算法能显著提高动态多目标优化方法的性能。

关键词(KeyWords)： 动态多目标优化;迁移学习;源域选择策略

Abstract:

Keywords:

基金项目(Foundation): 国家自然基金(61876123);; 山西省自然基金(201901D111262)

作者(Author): 上官晨曦,时振涛

参考文献(References)：

• [1] 白晓慧，何小娟，孙超利，等.基于分层学习的改进PSO算法求解复杂优化问题[J].太原科技大学学报，2021,42(3):169-174.
• [2] ZHOU J,ZOU J,YANG S,et al.An Evolutionary Dynamic Multi-objective Optimization Algorithm Based on Center-point Prediction and Sub-population Autonomous Guidance[C]// 2018 IEEE Symposium Series on Computational Intelligence (SSCI).IEEE,India,2019:2148-2154.
• [3] 刘若辰，李建霞，刘静，等.动态多目标优化研究综述 [J].计算机学报，2020,43(7):1247-1277.
• [4] RODRíGUEZ VILLALOBOS,CYNTHIA A,COELLO COELLO C A .A New Multi-Objective Evolutionary Algorithm Based on a Performance Assessment Indicator[C]//Conference on Genetic and Evolutionary Computation,Philly,USA,2012:505-512.
• [5] RUOCHEN,LIU,RUINAN,et al.Shape automatic clustering-based multi-objective optimization with decomposition[J].Machine Vision and Applications,2017,28(5-6):497-508.
• [6] LIANG H,FU W,YI F .A Survey of Recent Advances in Transfer Learning[C]//2019 IEEE 19th International Conference on Communication Technology (ICCT),Xi’an,China,2020:1516-1523.
• [7] LIN J,LIU H L,TAN K C,et al.An Effective Knowledge Transfer Approach for Multiobjective Multitasking Optimization[J].IEEE Transactions on Cybernetics,2020,99:1-11.
• [8] KUMAR S,SHUKLA A K,MUHURI P K,et al.Atanassov Intuitionistic Fuzzy Domain Adaptation to contain negative transfer learning[C]// 2016 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE,),Vancouver,BC,Canada,2016:2295-2301.
• [9] PAN S J,TSANG I W,KWOK J T,et al.Domain Adaptation via Transfer Component Analysis[J].IEEE Transactions on Neural Networks,2011,22(2):199-210.
• [10] PAN S J,KWOK J T,YANG Q .Transfer Learning via Dimensionality Reduction.[C]// Proceedings of the Twenty-Third AAAI Conference on Artificial Intelligence,Chicago,Illinois,USA,2008:677-682.
• [11] ISHIBUCHI H,MASUDA H,NOJIMA Y .Sensitivity of performance evaluation results by inverted generational distance to reference points[C]// 2016 IEEE Congress on Evolutionary Computation (CEC),Vancouver,BC,Canada,2016:1107-1114.
• [12] LAVANGNANANDA K,PHON-AMNUAISUK S,ENGCHUAN W,et al.Solving the IEEE CEC 2015 Dynamic Benchmark Problems Using Kalman Filter Based Dynamic Multiobjective Evolutionary Algorithm[J].Intelligent and Evolutionary Systems,Proceedings in Adaptation,Learning and Optimization 5,2016,DOI:10.1007/978-3-319-27000-5.
• [13] JIANG M,HUANG Z,QIU L,et al.Transfer Learning based Dynamic Multiobjective Optimization Algorithms[J].IEEE Transactions on Evolutionary Computation,2017,22(4):501-514.