融入任务空间转换和等分映射策略的多因子进化算法

doi:10.3778/j.issn.1002-8331.2412-0386

摘要/Abstract

摘要： 在运用多因子进化算法处理多任务优化问题时，不同任务之间的知识迁移可能会出现负迁移现象，以及算法容易陷入局部最优解等问题。为解决这些问题，提出了一种融入任务空间转换和等分映射策略的多因子进化算法（multifactorial evolutionary algorithm integrating task space transformation mechanism and equal-partitioning mapping strategy，MFEA-TSEM）。该算法通过引入任务空间转换来增强任务之间的相关性，从而促进任务之间的知识迁移。此外，所提出的等分映射策略应用于相同任务或不同任务之间的知识迁移，以避免任务陷入局部最优解并探索有希望的搜索区域。为了验证MFEA-TSEM算法的有效性，在单目标多任务优化问题和多目标多任务优化问题上与其他先进算法进行了比较。实验结果表明，MFEA-TSEM算法在保持解的多样性的同时，有效减少了负迁移现象的发生，从而提高了算法的全局搜索能力。

关键词: 进化算法, 多因子进化算法, 任务空间转换, 等分映射策略

Abstract: When applying multifactorial evolutionary algorithm to handle multi-task optimization problems, there may be negative transfer phenomena in the knowledge transfer between different tasks, and the algorithm is prone to getting trapped in local optimal solutions. To address these issues, a multifactorial evolutionary algorithm integrating task space transformation and equal-partitioning mapping strategy (MFEA-TSEM) is proposed. This algorithm enhances the correlation between tasks by introducing the task space transformation, thereby promoting knowledge transfer between tasks. In addition, the proposed equal-partitioning mapping strategy is applied to knowledge transfer within?the same task or between?different tasks to prevent tasks from getting trapped in local optima solutions and explore promising search regions. To verify the effectiveness of the MFEA-TSEM algorithm, it is compared with other advanced algorithms on single-objective multi-task optimization problems and multi-objective multi-task optimization problems. The experimental results show that the MFEA-TSEM algorithm effectively reduces the occurrence of negative transfer phenomena while maintaining the diversity of solutions, thus improving the global search ability of the algorithm.

Key words: evolutionary algorithm, multifactorial evolutionary algorithm, task space transformation, equal-partitioning mapping strategy

罗国星, 李治强, 刘飞龙, 李佩芸, 杨夏妮. 融入任务空间转换和等分映射策略的多因子进化算法[J]. 计算机工程与应用, 2025, 61(17): 185-199.

LUO Guoxing, LI Zhiqiang, LIU Feilong, LI Peiyun, YANG Xiani. Multifactorial Evolutionary Algorithm Integrating Task Space Transformation and Equal-Partitioning Mapping Strategy[J]. Computer Engineering and Applications, 2025, 61(17): 185-199.

参考文献

[1] GUPTA A, ONG Y S, FENG L. Multifactorial evolution: toward evolutionary multitasking[J]. IEEE Transactions on Evolutionary Computation, 2015, 20(3): 343-357.
[2] 李豪, 汪磊, 张元侨, 等. 演化多任务优化研究综述[J]. 软件学报, 2023, 34(2): 509-538.
LI H, WANG L, ZHANG Y Q, et al. Survey of evolutionary multitasking optimization[J]. Journal of Software, 2023, 34(2): 509-538.
[3] 李坚强, 蔡俊创, 孙涛, 等. 面向复杂物流配送场景的车辆路径规划多任务辅助进化算法[J]. 自动化学报, 2024, 50(3): 544-559.
LI J Q, CAI J C, SUN T, et al. Multitask-based assisted evolutionary algorithm for vehicle routing problems in complex logistics distribution scenarios[J]. Acta Automatica Sinica, 2024, 50(3): 544-559.
[4] FENG L, HUANG Y, ZHOU L, et al. Explicit evolutionary multitasking for combinatorial optimization: a case study on capacitated vehicle routing problem[J]. IEEE Transactions on Cybernetics, 2020, 51(6): 3143-3156.
[5] SHEN J, DONG H, TIAN Y, et al. Adaptive knowledge transfer based on machine learning method for evolutionary multitasking optimization[J]. Information Sciences, 2025, 702: 121908.
[6] LIU Z B, YUAN J H, ZHANG H L, et al. Optimal linear crossover for mitigating negative transfer in evolutionary multitasking[J]. IEEE Transactions on Evolutionary Computation, 2024, 99: 1.
[7] THANG T B, DAO T C, LONG N H, et al. Parameter adapt-ation in multifactorial evolutionary algorithm for many-task optimization[J]. Memetic Computing, 2021, 13(4): 433-446.
[8] BALI K K, ONG Y S, GUPTA A, et al. Multifactorial evolutionary algorithm with online transfer parameter estimation: MFEA-II[J]. IEEE Transactions on Evolutionary Computation, 2019, 24(1): 69-83.
[9] ZHOU L, FENG L, TAN K C, et al. Toward adaptive knowledge transfer in multifactorial evolutionary computation[J]. IEEE transactions on cybernetics, 2020, 51(5): 2563-2576.
[10] MA X, ZHENG Y, ZHU Z, et al. Improving evolutionary multitasking optimization by leveraging inter-task gene similarity and mirror transformation[J]. IEEE Computational Intelligence Magazine, 2021, 16(4): 38-53.
[11] WEI T, WANG S, ZHONG J, et al. A review on evolutionary multitask optimization: trends and challenges[J]. IEEE Transactions on Evolutionary Computation, 2021, 26(5): 941-960.
[12] WANG Z Z, CAO L L, FENG L, et al. Evolutionary multitask optimization with lower confidence bound-based solution selection strategy[J]. IEEE Transactions on Evolutionary Computation, 2025, 29(1): 132-144.
[13] FENG L, ZHOU L, ZHONG J, et al. Evolutionary multitasking via explicit autoencoding[J]. IEEE transactions on cybernetics, 2018, 49(9): 3457-3470.
[14] DING J, YANG C, JIN Y, et al. Generalized multitasking for evolutionary optimization of expensive problems[J]. IEEE Transactions on Evolutionary Computation, 2017, 23(1): 44-58.
[15] BALI K K, GUPTA A, FENG L, et al. Linearized domain adaptation in evolutionary multitasking[C]//Proceedings of the 2017 IEEE Congress on Evolutionary Computation (CEC). New York: ACM, 2017: 1295-1302.
[16] XUE X, ZHANG K, TAN K C, et al. Affine transformation-enhanced multifactorial optimization for heterogeneous problems[J]. IEEE Transactions on Cybernetics, 2020, 52(7): 6217-6231.
[17] GUPTA A, ONG Y S, FENG L. Insights on transfer optimization: because experience is the best teacher[J]. IEEE Transactions on Emerging Topics in Computational Intelligence, 2017, 2(1): 51-64.
[18] STEWART G W. On the early history of the singular value decomposition[J]. SIAM Review, 1993, 35(4): 551-566.
[19] HU Z Y, WANG S, LI Y L, et al. Optimisation of steel rol-ling schedule based on evolutionary multi-tasking transfer algorithm[J]. Computers & Operations Research, 2024, 169: 106743.
[20] ZHANG T, LI D, Li Y, et al. Constrained multitasking optimization via co-evolution and domain adaptation[J]. Swarm and Evolutionary Computation, 2024, 87: 101570.
[21] JAIN H, DEB K. An evolutionary many-objective optimization algorithm using reference-point based nondominated sorting approach, part II: handling constraints and exten-ding to an adaptive approach[J]. IEEE Transactions on Evolutionary Computation, 2013, 18(4): 602-622.
[22] HOTELLING H. Relations between two sets of variates[M]//Breakthroughs in statistics: methodology and distribution. New York, NY: Springer New York, 1992: 162-190.
[23] LIM R, GUPTA A, ONG Y S, et al. Non-linear domain adapt-ation in transfer evolutionary optimization[J]. Cognitive Computation, 2021, 13: 290-307.
[24] DEB K, BEYER H G. Self-adaptive genetic algorithms with simulated binary crossover[J]. Evolutionary Computation, 2001, 9(2): 197-221.
[25] 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.
[26] DA B, ONG Y S, FENG L, et al. Evolutionary multitasking for single-objective continuous optimization: benchmark problems, performance metric, and baseline results[J]. arXiv:1706. 03470, 2017.
[27] YUAN Y, ONG Y S, FENG L, et al. Evolutionary multitas-king for multiobjective continuous optimization: benchmark problems, performance metrics and baseline results[J]. arXiv：1706.02766, 2017.
[28] ZAR J H. Significance testing of the Spearman rank correl-ation coefficient[J]. Journal of the American Statistical Association, 1972, 67: 578-580.
[29] FENG L, ZHOU W, ZHOU L, et al. An empirical study of multifactorial PSO and multifactorial DE[C]//Proceedings of the 2017 IEEE Congress on Evolutionary Computation. Piscataway: IEEE, 2017: 921-928.
[30] CHEN A, REN Z, WANG M, et al. Aligning heterogeneous optimization problems with optimal correspondence assisted affine transformation for evolutionary multi-tasking[J]. Applied Soft Computing, 2023, 136: 110070.
[31] LIU Z B, LI G, ZHANG H L, et al. Multifactorial evolutionary algorithm based on diffusion gradient descent[J]. IEEE Transactions on Cybernetics, 2024, 54(7): 4267-4279.
[32] GUPTA A, ONG Y S, FENG L, et al. Multiobjective multifactorial optimization in evolutionary multitasking[J]. IEEE Transactions on Cybernetics, 2016, 47(7): 1652-1665.
[33] BALI K K, GUPTA A, ONG Y S, et al. Cognizant multitas-king in multiobjective multifactorial evolution: MO-MFEA-II[J]. IEEE Transactions on Cybernetics, 2020, 51(4): 1784-1796.
[34] YANG C E, DING J L, TAN K C, et al. Two-stage assortative mating for multi-objective multifactorial evolutionary optimization[C]//Proceedings of the 2017 IEEE 56th Annual Conference on Decision and Control (CDC). New York: ACM, 2017: 76-81.
[35] ZITZLER E, THIELE L, LAUMANNS M, et al. Performance assessment of multiobjective optimizers: an analysis and review[J]. IEEE Transactions on evolutionary comput-ation, 2003, 7(2): 117-132.
[36] CARRASCO J, GARCíA S, RUEDA M M, et al. Recent trends in the use of statistical tests for comparing swarm and evolutionary computing algorithms: practical guidelines and a critical review[J]. Swarm and Evolutionary Comput-ation, 2020, 54: 100665.
[37] CHEN Y, ZHONG J, FENG L, et al. An adaptive archive-based evolutionary framework for many-task optimization[J]. IEEE Transactions on Emerging Topics in Computational Intelligence, 2019, 4(3): 369-384.