Multitask differential evolution optimization based on an expert library module

Yuehui Zhang; Xuewen Xia; Yi Zeng; Fenglin Lin; Yuehui Zhang; Xuewen Xia; Yi Zeng; Fenglin Lin

doi:10.3934/era.2026189

Electronic Research Archive

2026, Volume 34, Issue 6: 4216-4247. doi: 10.3934/era.2026189

Previous Article Next Article

Research article

Multitask differential evolution optimization based on an expert library module

1.
College of Physics and Information Engineering, Minnan Normal University, Zhangzhou 363000, China
2.
Key Lab of Intelligent Optimization and Information Processing, Minnan Normal University, Zhangzhou 363000, China

Received: 06 March 2026 Revised: 03 May 2026 Accepted: 12 May 2026 Published: 19 May 2026

Utilizing evolutionary algorithms (EAs) to solve multitask optimization problems (MTOPs) simultaneously has gained significant attention because many real applications usually share similar properties or have a certain degree of correlations. In this new research area in the EA community, namely multitask optimization (MTO), a key challenge is transfering useful knowledge among different tasks efficiently. In this paper, a multitask differential evolution (DE) algorithm is proposed in which a simple variant of the canonical DE is adopted as the basic optimizer. Furthermore, an expert library module (ELM), which can extract useful knowledge from the historical optimization experiences of DE, is introduced as a complement for the DE to generate offspring. In the proposed algorithm, named MTDE-ELM, when the DE performs the mutation operator, if all individuals chosen to perform the mutation have the same skill factor, intratask knowledge transfer can be realized. Conversely, when individuals with different skill factors are adopted to execute the mutation, intertask knowledge transfer can be realized. In the ELM, evolutionary data of different tasks are used to train the feedforward neural networks (FNNs). After that, the trained FNN can realize potential knowledge transfer among different tasks. Moreover, based on the trained FNN, each individual can query its promising evolutionary direction, which can be regarded as an effective supplement for the DE to generate offspring. The comprehensive performance of the MTDE-ELM is demonstrated through comparisons with five other MTO algorithms. Furthermore, distinct properties of the newly introduced strategies are also confirmed by experimental analysis.
- evolutionary algorithms,
- multitask optimization,
- knowledge transfer,
- differential evolution,
- feedforward neural network
Citation: Yuehui Zhang, Xuewen Xia, Yi Zeng, Fenglin Lin. Multitask differential evolution optimization based on an expert library module[J]. Electronic Research Archive, 2026, 34(6): 4216-4247. doi: 10.3934/era.2026189

Related Papers:

Abstract

Utilizing evolutionary algorithms (EAs) to solve multitask optimization problems (MTOPs) simultaneously has gained significant attention because many real applications usually share similar properties or have a certain degree of correlations. In this new research area in the EA community, namely multitask optimization (MTO), a key challenge is transfering useful knowledge among different tasks efficiently. In this paper, a multitask differential evolution (DE) algorithm is proposed in which a simple variant of the canonical DE is adopted as the basic optimizer. Furthermore, an expert library module (ELM), which can extract useful knowledge from the historical optimization experiences of DE, is introduced as a complement for the DE to generate offspring. In the proposed algorithm, named MTDE-ELM, when the DE performs the mutation operator, if all individuals chosen to perform the mutation have the same skill factor, intratask knowledge transfer can be realized. Conversely, when individuals with different skill factors are adopted to execute the mutation, intertask knowledge transfer can be realized. In the ELM, evolutionary data of different tasks are used to train the feedforward neural networks (FNNs). After that, the trained FNN can realize potential knowledge transfer among different tasks. Moreover, based on the trained FNN, each individual can query its promising evolutionary direction, which can be regarded as an effective supplement for the DE to generate offspring. The comprehensive performance of the MTDE-ELM is demonstrated through comparisons with five other MTO algorithms. Furthermore, distinct properties of the newly introduced strategies are also confirmed by experimental analysis.

References

[1]	S. Han, W. Pedrycz, C. Han, Nonlinear channel blind equalization using hybrid genetic algorithm with simulated annealing, Math. Comput. Modell., 41 (2005), 697–709. https://doi.org/10.1016/j.mcm.2004.05.006 doi: 10.1016/j.mcm.2004.05.006
[2]	T. Back, U. Hammel, H. P. Schwefel, Evolutionary computation: Comments on the history and current state, IEEE Trans. Evol. Comput., 1 (1997), 3–17. https://doi.org/10.1109/4235.585888 doi: 10.1109/4235.585888
[3]	S. Sen, S. Agarwal, P. Chakraborty, K. P. Singh, Astronomical big data processing using machine learning: A comprehensive review, Exp. Astron., 53 (2022), 1–43. https://doi.org/10.1007/s10686-021-09827-4 doi: 10.1007/s10686-021-09827-4
[4]	P. Charbonneau, Genetic algorithms in astronomy and astrophysics, Astrophys. J. Suppl. Ser., 101 (1995), 309–334. https://doi.org/10.1086/192242 doi: 10.1086/192242
[5]	J. Caspermeyer, Researchers show nature conserves its most vital dna by multitasking, Mol. Biol. Evol., 33 (2016), 2477–2478. https://doi.org/10.1093/molbev/msw113 doi: 10.1093/molbev/msw113
[6]	X. Zhou, A. K. Qin, M. Gong, K. C. Tan, A survey on evolutionary construction of deep neural networks, IEEE Trans. Evol. Comput., 25 (2021), 894–912. https://doi.org/10.1109/TEVC.2021.3079985 doi: 10.1109/TEVC.2021.3079985
[7]	L. Wang, J. Li, X. Yan, A variable population size opposition-based learning for differential evolution algorithm and its applications on feature selection, Appl. Intell., 54 (2024), 959–984. https://doi.org/10.1007/s10489-023-05179-y doi: 10.1007/s10489-023-05179-y
[8]	T. Wei, S. Wang, J. Zhong, D. Liu, J. Zhang, A review on evolutionary multitask optimization: Trends and challenges, IEEE Trans. Evol. Comput., 26 (2021), 941–960. https://doi.org/10.1109/TEVC.2021.3139437 doi: 10.1109/TEVC.2021.3139437
[9]	A. Gupta, Y. S. Ong, L. Feng, K. C. Tan, Multiobjective multifactorial optimization in evolutionary multitasking, IEEE Trans. Cybern., 47 (2016), 1652–1665. https://doi.org/10.1109/TCYB.2016.2554622 doi: 10.1109/TCYB.2016.2554622
[10]	J. Ding, C. Yang, Y. Jin, T. Chai, Generalized multitasking for evolutionary optimization of expensive problems, IEEE Trans. Evol. Comput., 23 (2019), 44–58. https://doi.org/10.1109/TEVC.2017.2785351 doi: 10.1109/TEVC.2017.2785351
[11]	Z. G. Chen, Z. H. Zhan, S. Kwong, J. Zhang, Evolutionary computation for intelligent transportation in smart cities: A survey, IEEE Comput. Intell. Mag., 17 (2022), 83–102. https://doi.org/10.1109/MCI.2022.3155330 doi: 10.1109/MCI.2022.3155330
[12]	K. C. Tan, L. Feng, M. Jiang, Evolutionary transfer optimization-a new frontier in evolutionary computation research, IEEE Comput. Intell. Mag., 16 (2021), 22–33. https://doi.org/10.1109/MCI.2020.3039066 doi: 10.1109/MCI.2020.3039066
[13]	Y. Wang, Q. Zhang, G. G. Wang, Improving evolutionary algorithms with information feedback model for large-scale many-objective optimization, Appl. Intell., 53 (2023), 11439–11473. https://doi.org/10.1007/s10489-022-03964-9 doi: 10.1007/s10489-022-03964-9
[14]	H. Zhao, X. Ning, X. Liu, C. Wang, J. Liu, What makes evolutionary multi-task optimization better: A comprehensive survey, Appl. Soft Comput., 145 (2023), 110545. https://doi.org/10.1016/j.asoc.2023.110545 doi: 10.1016/j.asoc.2023.110545
[15]	W. Yang, J. Liu, S. Tan, W. Zhang, Y. Liu, Evolutionary dynamic grouping based cooperative co-evolution algorithm for large-scale optimization, Appl. Intell., 54 (2024), 4585–4601. https://doi.org/10.1007/s10489-024-05390-5 doi: 10.1007/s10489-024-05390-5
[16]	X. Zheng, A. K. Qin, M. Gong, D. Zhou, Self-regulated evolutionary multitask optimization, IEEE Trans. Evol. Comput., 24 (2019), 16–28. https://doi.org/10.1109/TEVC.2019.2904696 doi: 10.1109/TEVC.2019.2904696
[17]	X. Xia, L. Gui, F. Yu, H. Wu, B. Wei, Y. L. Zhang, et al., Triple archives particle swarm optimization, IEEE Trans. Cybern., 50 (2019), 4862–4875. https://doi.org/10.1109/TCYB.2019.2943928 doi: 10.1109/TCYB.2019.2943928
[18]	F. Gao, L. Huang, W. Gao, L. Li, S. Wang, M. Gong, et al., Transferring knowledge by budget online learning for multiobjective multitasking optimization, Swarm Evol. Comput., 91 (2024), 101765. https://doi.org/10.1016/j.swevo.2024.101765 doi: 10.1016/j.swevo.2024.101765
[19]	Y. Guo, G. Chen, M. Jiang, D. Gong, J. Liang, A knowledge guided transfer strategy for evolutionary dynamic multiobjective optimization, IEEE Trans. Evol. Comput., 27 (2022), 1750–1764. https://doi.org/10.1109/TEVC.2022.3222844 doi: 10.1109/TEVC.2022.3222844
[20]	Z. Cui, B. Zhao, T. Zhao, X. Cai, J. Chen, Adaptive multi-task evolutionary algorithm based on knowledge reuse, Inf. Sci., 648 (2023), 119568. https://doi.org/10.1016/j.ins.2023.119568 doi: 10.1016/j.ins.2023.119568
[21]	A. Gupta, Y. Ong, L. Feng, Multifactorial evolution: Toward evolutionary multitasking, IEEE Trans. Evol. Comput., 20 (2016), 343–357. https://doi.org/10.1109/TEVC.2015.2458037 doi: 10.1109/TEVC.2015.2458037
[22]	L. Zhou, L. Feng, K. C. Tan, J. Zhong, Z. Zhu, K. Liu, et al., Toward adaptive knowledge transfer in multifactorial evolutionary computation, IEEE Trans. Cybern., 51 (2021), 2563–2576. https://doi.org/10.1109/TCYB.2020.2974100 doi: 10.1109/TCYB.2020.2974100
[23]	Z. H. Zhan, J. Y. Li, S. Kwong, J. Zhang, Learning-aided evolution for optimization, IEEE Trans. Evol. Comput., 27 (2022), 1794–1808. https://doi.org/10.1109/TEVC.2022.3232776 doi: 10.1109/TEVC.2022.3232776
[24]	J. Y. Li, Z. H. Zhan, K. C. Tan, J. Zhang, A meta-knowledge transfer-based differential evolution for multitask optimization, IEEE Trans. Evol. Comput., 26 (2021), 719–734. https://doi.org/10.1109/TEVC.2021.3131236 doi: 10.1109/TEVC.2021.3131236
[25]	R. Chandra, A. Gupta, Y. S. Ong, C. K. Goh, Evolutionary multi-task learning for modular knowledge representation in neural networks, Neural Process. Lett., 47 (2018), 993–1009. https://doi.org/10.1007/s11063-017-9718-z doi: 10.1007/s11063-017-9718-z
[26]	C. Qin, T. Zhu, K. Jiang, Y. Wu, J. Zhang, Neural-network-based safe learning control for non-zero-sum differential games of nonlinear systems with asymmetric input constraints, Appl. Intell., 54 (2024), 7810–7828. https://doi.org/10.1007/s10489-024-05593-w doi: 10.1007/s10489-024-05593-w
[27]	Z. F. Xue, Z. J. Wang, Z. H. Zhan, S. Kwong, J. Zhang, Neural network-based knowledge transfer for multitask optimization, IEEE Trans. Cybern., 54 (2024), 7541–7554. https://doi.org/10.1109/TCYB.2024.3469371 doi: 10.1109/TCYB.2024.3469371
[28]	Y. Jiang, Z. H. Zhan, K. C. Tan, J. Zhang, Knowledge learning for evolutionary computation, IEEE Trans. Evol. Comput., 29 (2023), 16–30. https://doi.org/10.1109/TEVC.2023.3278132 doi: 10.1109/TEVC.2023.3278132
[29]	Q. Rui, W. L. Liu, Y. Wu, J. Zhong, Nonlinear mapping meets multi-task bayesian optimization: A knowledge transfer perspective, in 2025 IEEE International Conference on Systems, Man, and Cybernetics (SMC), (2025), 1414–1419. https://doi.org/10.1109/SMC58881.2025.11343035
[30]	J. Guo, L. Li, H. Sun, M. Qin, H. Yu, T. Zhang, A min-max optimization framework for sparse multi-task deep neural network, Neurocomputing, 650 (2025), 130865. https://doi.org/10.1016/j.neucom.2025.130865 doi: 10.1016/j.neucom.2025.130865
[31]	H. Han, B. Zhao, X. Wu, X. Li, Survey on multi-task optimization: Towards cross-domain and asynchronous multi-task, Swarm Evol. Comput., 99 (2025), 102175. https://doi.org/10.1016/j.swevo.2025.102175 doi: 10.1016/j.swevo.2025.102175
[32]	C. Wang, L. Li, L. Jiao, J. Zhao, F. Liu, S. Yang, Learning evolution via optimization knowledge adaptation, IEEE Trans. Pattern Anal. Mach. Intell., 2026 (2026), 1–18. https://doi.org/10.1109/TPAMI.2026.3686919 doi: 10.1109/TPAMI.2026.3686919
[33]	R. Storn, K. Price, Differential evolution–a simple and efficient heuristic for global optimization over continuous spaces, J. Glob. Optim., 11 (1997), 341–359. https://doi.org/10.1023/A:1008202821328 doi: 10.1023/A:1008202821328
[34]	L. Feng, Y. S. Ong, S. Jiang, A. Gupta, Autoencoding evolutionary search with learning across heterogeneous problems, IEEE Trans. Evol. Comput., 21 (2017), 760–772. https://doi.org/10.1109/TEVC.2017.2682274 doi: 10.1109/TEVC.2017.2682274
[35]	H. Li, Y. S. Ong, M. Gong, Z. Wang, Evolutionary multitasking sparse reconstruction: Framework and case study, IEEE Trans. Evol. Comput., 23 (2018), 733–747. https://doi.org/10.1109/TEVC.2018.2881955 doi: 10.1109/TEVC.2018.2881955
[36]	K. Bali, Y. Ong, A. Gupta, P. S. Tan, Multifactorial evolutionary algorithm with online transfer parameter estimation: Mfea-ii, IEEE Trans. Evol. Comput., 24 (2020), 69–83. https://doi.org/10.1109/TEVC.2019.2906927 doi: 10.1109/TEVC.2019.2906927
[37]	Z. Liang, J. Zhang, L. Feng, Z. Zhu, A hybrid of genetic transform and hyper-rectangle search strategies for evolutionary multi-tasking, Exp. Syst. Appl., 138 (2019), 112798. https://doi.org/10.1016/j.eswa.2019.07.015 doi: 10.1016/j.eswa.2019.07.015
[38]	M. Gong, Z. Tang, H. Li, J. Zhang, Evolutionary multitasking with dynamic resource allocating strategy, IEEE Trans. Evol. Comput., 23 (2019), 858–869. https://doi.org/10.1109/TEVC.2019.2893614 doi: 10.1109/TEVC.2019.2893614
[39]	C. Wang, J. Liu, K. Wu, Z. Wu, Solving multitask optimization problems with adaptive knowledge transfer via anomaly detection, IEEE Trans. Evol. Comput., 26 (2021), 304–318. https://doi.org/10.1109/TEVC.2021.3068157 doi: 10.1109/TEVC.2021.3068157
[40]	K. K. Bali, A. Gupta, L. Feng, Y. S. Ong, Linearized domain adaptation in evolutionary multitasking, in 2017 IEEE Congress on Evolutionary Computation (CEC), (2017), 1295–1302. https://doi.org/10.1109/CEC.2017.7969454
[41]	X. Wang, Q. Kang, M. Zhou, S. Yao, A. Abusorrah, Domain adaptation multitask optimization, IEEE Trans. Cybern., 53 (2023), 4567–4578. https://doi.org/10.1109/TCYB.2022.3222101 doi: 10.1109/TCYB.2022.3222101
[42]	H. Han, X. Bai, Y. Hou, J. Qiao, Multitask particle swarm optimization with heterogeneous domain adaptation, IEEE Trans. Evol. Comput., 28 (2024), 178–192. https://doi.org/10.1109/TEVC.2023.3258491 doi: 10.1109/TEVC.2023.3258491
[43]	X. Xia, L. Gui, Y. Zhang, X. Xu, F. Yu, A fitness-based adaptive differential evolution algorithm, Inf. Sci., 549 (2021), 116–141. https://doi.org/10.1016/j.ins.2020.11.015 doi: 10.1016/j.ins.2020.11.015
[44]	Z. Liang, X. Xu, L. Liu, Y. Tu, Z. Zhu, Evolutionary many-task optimization based on multisource knowledge transfer, IEEE Trans. Evol. Comput., 26 (2022), 319–333. https://doi.org/10.1109/TEVC.2021.3101697 doi: 10.1109/TEVC.2021.3101697
[45]	S. H. Wu, Z. H. Zhan, K. C. Tan, J. Zhang, Transferable adaptive differential evolution for many-task optimization, IEEE Trans. Cybern., 53 (2023), 7295–7308. https://doi.org/10.1109/TCYB.2023.3234969 doi: 10.1109/TCYB.2023.3234969
[46]	T. Zhang, W. Gong, Y. Li, Multitask differential evolution with adaptive dual knowledge transfer, Appl. Soft Comput., 165 (2024), 112040. https://doi.org/10.1016/j.asoc.2024.112040 doi: 10.1016/j.asoc.2024.112040
[47]	X. Ma, M. Xu, Y. Yu, H. Liu, Y. Wang, L. Wang, et al., Enhancing evolutionary multitasking optimization by leveraging intertask knowledge transfers and improved evolutionary operators, Knowl. Based Syst., 259 (2023), 110027. https://doi.org/10.1016/j.knosys.2022.110027 doi: 10.1016/j.knosys.2022.110027
[48]	L. Feng, W. Zhou, L. Zhou, S. W. Jiang, J. H. Zhong, B. S. Da, et al., An empirical study of multifactorial PSO and multifactorial DE, in IEEE Congress on Evolutionary Computation (CEC), (2017), 921–928. https://doi.org/10.1109/CEC.2017.7969407
[49]	Z. Tang, M. Gong, Y. Wu, W. Liu, Y. Xie, Regularized evolutionary multitask optimization: Learning to intertask transfer in aligned subspace, IEEE Trans. Evol. Comput., 25 (2021), 262–276. https://doi.org/10.1109/TEVC.2020.3023480 doi: 10.1109/TEVC.2020.3023480
[50]	G. Yuan, G. Sun, L. Deng, C. Li, G. Yang, A novel differential evolution algorithm based on periodic intervention and systematic regulation mechanisms, Appl. Intell., 54 (2024), 11779–11803. https://doi.org/10.1007/s10489-024-05781-8 doi: 10.1007/s10489-024-05781-8
[51]	J. H. Holland, Genetic algorithms, Sci. Am., 267 (1992), 66–73. https://doi.org/10.1038/scientificamerican0792-66
[52]	J. Liu, L. Tong, X. Xia, A genetic algorithm for vehicle routing problems with time windows based on cluster of geographic positions and time windows, Appl. Soft Comput., 169 (2025), 112593. https://doi.org/10.1016/j.asoc.2024.112593 doi: 10.1016/j.asoc.2024.112593
[53]	A. Sureyya Rifaioglu, T. Doğan, M. Jesus Martin, R. Cetin-Atalay, V. Atalay, Deepred: Automated protein function prediction with multi-task feed-forward deep neural networks, Sci. Rep., 9 (2019), 7344. https://doi.org/10.1038/s41598-019-43708-3 doi: 10.1038/s41598-019-43708-3
[54]	M. Moller, Efficient training of feed-forward neural networks, in Neural Network Analysis, Architectures and Applications, CRC Press, (2024), 136–173. https://doi.org/10.1201/9781003572886-8
[55]	D. E. Rumelhart, G. E. Hinton, R. J. Williams, Learning representations by back-propagating errors, Nature, 323 (1986), 533–536. https://doi.org/10.1038/323533a0 doi: 10.1038/323533a0
[56]	S. Schmidgall, R. Ziaei, J. Achterberg, L. Kirsch, S. Hajiseyedrazi, J. Eshraghian, Brain-inspired learning in artificial neural networks: A review, APL Mach. Learn., 2 (2024), 021501. https://doi.org/10.1063/5.0186054 doi: 10.1063/5.0186054
[57]	Y. LeCun, Y. Bengio, G. Hinton, Deep learning, Nature, 521 (2015), 436–444. https://doi.org/10.1038/nature14539
[58]	B. Da, Y. S. Ong, K. C. Tan, A. K. Qin, A. Gupta, Z. Zhu, et al., Evolutionary multitasking for single-objective continuous optimization: Benchmark problems, performance metric, and baseline results, preprint, arXiv: 1706.03470. https://doi.org/10.48550/arXiv.1706.03470
[59]	Y. Yuan, Y. S. Ong, L. Feng, A. K. Qin, A. Gupta, B. Da, et al., Evolutionary multitasking for multiobjective continuous optimization: Benchmark problems, performance metrics and baseline results, preprint, arXiv: 1706.02766. https://doi.org/10.48550/arXiv.1706.02766
[60]	L. Feng, L. Zhou, J. Zhong, A. Gupta, Y. S. Ong, K. C. Tan, et al., Evolutionary multitasking via explicit autoencoding, IEEE Trans. Cybern., 49 (2019), 3457–3470. https://doi.org/10.1109/TCYB.2018.2845361 doi: 10.1109/TCYB.2018.2845361
[61]	Y. Li, W. Gong, S. Li, Multitasking optimization via an adaptive solver multitasking evolutionary framework, Inf. Sci., 630 (2023), 688–712. https://doi.org/10.1016/j.ins.2022.10.099 doi: 10.1016/j.ins.2022.10.099
[62]	Y. Jiang, Z. H. Zhan, K. C. Tan, J. Zhang, Block-level knowledge transfer for evolutionary multitask optimization, IEEE Trans. Cybern., 54 (2024), 558–571. https://doi.org/10.1109/TCYB.2023.3273625 doi: 10.1109/TCYB.2023.3273625
[63]	Y. Li, W. Gong, Multiobjective multitask optimization with multiple knowledge types and transfer adaptation, IEEE Trans. Evol. Comput., 29 (2025), 205–216. https://doi.org/10.1109/TEVC.2024.3353319 doi: 10.1109/TEVC.2024.3353319
[64]	K. Deb, R. B. Agrawal, Simulated binary crossover for continuous search space, Complex Syst., 9 (1995), 115–148.
[65]	K. Deb, M. Goyal, A combined genetic adaptive search (geneas) for engineering design, Comput. Sci. Inf. Syst, 26 (1996), 30–45.
[66]	J. Carrasco, S. García, M. Rueda, S. Das, F. Herrera, Recent trends in the use of statistical tests for comparing swarm and evolutionary computing algorithms: Practical guidelines and a critical review, Swarm Evol. Comput., 54 (2020), 100665. https://doi.org/10.1016/j.swevo.2020.100665 doi: 10.1016/j.swevo.2020.100665

Reader Comments

Your name:*

Email:*
© 2026 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)