A multitask optimization algorithm based on elite individual transfer

Yutao Lai; Hongyan Chen; Fangqing Gu; Yutao Lai; Hongyan Chen; Fangqing Gu

doi:10.3934/mbe.2023360

Mathematical Biosciences and Engineering

2023, Volume 20, Issue 5: 8261-8278. doi: 10.3934/mbe.2023360

Previous Article Next Article

Research article Special Issues

A multitask optimization algorithm based on elite individual transfer

School of Mathematics and Statistics, Guangdong University of Technology, Guangzhou, China

Academic Editor: Yiu-ming Cheung

Received: 13 January 2023 Revised: 14 February 2023 Accepted: 21 February 2023 Published: 28 February 2023

Evolutionary multitasking algorithms aim to solve several optimization tasks simultaneously, and they can improve the efficiency of various tasks evolution through the knowledge transfer between different optimization tasks. Evolutionary multitasking algorithms have been applied to various applications and achieved certain results. However, how to transfer knowledge between tasks is still a problem worthy of research. Aiming to improve the positive transfer between tasks and reduce the negative transfer, we propose a single-objective multitask optimization algorithm based on elite individual transfer, namely MSOET. In this paper, whether to execute knowledge transfer between tasks depends on a certain probability. Meanwhile, in order to enhance the effectiveness and the global search ability of the algorithm, the current population and the elite individual in the transfer population are further utilized as the learning sources to construct a Gaussian distribution model, and the offspring is generated by the Gaussian distribution model to achieve knowledge transfer between tasks. We compared the proposed MSOET with ten multitask optimization algorithms, and the experimental results verify the algorithm's excellent performance and strong robustness.
- multitask optimization,
- Gaussian distribution,
- knowledge transfer,
- evolutionary algorithm,
- evolutionary multitasking algorithms
Citation: Yutao Lai, Hongyan Chen, Fangqing Gu. A multitask optimization algorithm based on elite individual transfer[J]. Mathematical Biosciences and Engineering, 2023, 20(5): 8261-8278. doi: 10.3934/mbe.2023360

Related Papers:

Abstract

Evolutionary multitasking algorithms aim to solve several optimization tasks simultaneously, and they can improve the efficiency of various tasks evolution through the knowledge transfer between different optimization tasks. Evolutionary multitasking algorithms have been applied to various applications and achieved certain results. However, how to transfer knowledge between tasks is still a problem worthy of research. Aiming to improve the positive transfer between tasks and reduce the negative transfer, we propose a single-objective multitask optimization algorithm based on elite individual transfer, namely MSOET. In this paper, whether to execute knowledge transfer between tasks depends on a certain probability. Meanwhile, in order to enhance the effectiveness and the global search ability of the algorithm, the current population and the elite individual in the transfer population are further utilized as the learning sources to construct a Gaussian distribution model, and the offspring is generated by the Gaussian distribution model to achieve knowledge transfer between tasks. We compared the proposed MSOET with ten multitask optimization algorithms, and the experimental results verify the algorithm's excellent performance and strong robustness.

References

[1]	K. K. Bali, Y. S. 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
[2]	R. Liaw, C. Ting, Evolutionary many-tasking based on biocoenosis through symbiosis: A framework and benchmark problems, in 2017 IEEE Congress on Evolutionary Computation, CEC 2017, Donostia, (2017), 2266–2273. https://doi.org/10.1109/cec.2017.7969579
[3]	X. Ma, J. Yin, A. Zhu, X. Li, Y. Yu, L. Wang, et al., Enhanced multifactorial evolutionary algorithm with meme helper-tasks, IEEE Trans. Cybern., 52 (2022), 7837–7851. https://doi.org/10.1109/tcyb.2021.3050516 doi: 10.1109/tcyb.2021.3050516
[4]	H. L. Liu, F. Gu, Y. M. Cheung, S. Xie, J. Zhang, On solving WCDMA network planning using iterative power control scheme and evolutionary multiobjective algorithm, IEEE Comput. Intell. Mag., 9 (2014), 44–52. https://doi.org/10.1109/mci.2013.2291690 doi: 10.1109/mci.2013.2291690
[5]	J. Yi, W. Zhang, J. Bai, W. Zhou, L. Yao, Multifactorial evolutionary algorithm based on improved dynamical decomposition for many-objective optimization problems, IEEE Trans. Evol. Comput., 26 (2022), 334–348. https://doi.org/10.1109/tevc.2021.3135691 doi: 10.1109/tevc.2021.3135691
[6]	L. Chen, H. L. Liu, K. C. Tan, Y. M. Cheung, Y. Wang, Evolutionary many-objective algorithm using decomposition-based dominance relationship, IEEE Trans. Cybern., 49 (2019), 4129–4139. https://doi.org/10.1109/TCYB.2018.2859171 doi: 10.1109/TCYB.2018.2859171
[7]	H. L. Liu, L. Chen, Q. Zhang, K. Deb, Evolutionary many-objective algorithm using decomposition-based dominance relationship, IEEE Trans. Evol. Comput., 22 (2018), 433–448. https://doi.org/10.1109/TEVC.2017.2725902 doi: 10.1109/TEVC.2017.2725902
[8]	Q. Peng, Y. M. Cheung, X. You, Y. Y. Tang, A hybrid of local and global saliencies for detecting image salient region and appearance, IEEE Trans. Syst. Man Cybern. Syst., 47 (2017), 86–97. https://doi.org/10.1109/TSMC.2016.2564922 doi: 10.1109/TSMC.2016.2564922
[9]	F. Zhang, Y. Mei, S. Nguyen, M. Zhang, Multitask multiobjective genetic programming for automated scheduling heuristic learning in dynamic flexible job-shop scheduling, IEEE Trans. Cybern., 2022 (2022), 1–14. https://doi.org/10.1109/tcyb.2022.3196887 doi: 10.1109/tcyb.2022.3196887
[10]	S. Tuo, C. Li, F. Liu, A. Li, L. He, Z. W. Geem, et al., MTHSA-DHEI: multitasking harmony search algorithm for detecting high-order SNP epistatic interactions, Complex Intell. Syst., 9 (2023), 637–658. https://doi.org/10.1007/s40747-022-00813-7 doi: 10.1007/s40747-022-00813-7
[11]	J. Zhong, L. Feng, W. Cai, Y. S. Ong, Multifactorial genetic programming for symbolic regression problems, IEEE Trans. Syst. Man Cybern. Syst., 50 (2020), 4492–4505. https://doi.org/10.1109/tsmc.2018.2853719 doi: 10.1109/tsmc.2018.2853719
[12]	S. Handoko, H. Lau, A. Gupta, Y. Ong, H. Kim, P. Tan, Solving multi-vehicle profitable tour problem via knowledge adoption in evolutionary bi-level programming, in 2022 IEEE Congress on Evolutionary Computation (CEC), (2015), 2713–2720. https://doi.org/10.1109/cec.2015.7257225
[13]	F. Gu, H. L. Liu, Y. M. Cheung, M. Zheng, A rough-to-fine evolutionary multiobjective optimization algorithm, IEEE Trans. Cybern., 52 (2022), 13472–13485. https://doi.org/10.1109/tcyb.2021.3081357 doi: 10.1109/tcyb.2021.3081357
[14]	R. Geng, R. Ji, S. Zi, Research on task allocation of UAV cluster based on particle swarm quantization algorithm, Math. Biosci. Eng., 20 (2022), 18–33. https://doi.org/10.3934/mbe.2023002 doi: 10.3934/mbe.2023002
[15]	A. Gupta, J. Mańdziuk, Y. S. Ong, Evolutionary multitasking in bi-level optimization, Complex Intell. Syst., 1 (2015), 83–95. https://doi.org/10.1007/s40747-016-0011-y doi: 10.1007/s40747-016-0011-y
[16]	M. Xu, Y. Zheng, Y. S. Ong, Z. Zhu, X. Ma, A multifactorial differential evolution with hybrid global and local search strategies, in 2022 IEEE Congress on Evolutionary Computation (CEC), (2022), 1–9. https://doi.org/10.1109/cec55065.2022.9870335
[17]	S. J. Pan, Q. Yang, A survey on transfer learning, IEEE Trans. Knowl. Data Eng., 22 (2010), 1345–1359. https://doi.org/10.5220/0006396700170027 doi: 10.5220/0006396700170027
[18]	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
[19]	A. Gupta, Y. S. 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
[20]	Z. Liang, H. Dong, C. Liu, W. Liang, Z. Zhu, Evolutionary multitasking for multiobjective optimization with subspace alignment and adaptive differential evolution, IEEE Trans. Cybern., 52 (2022), 2096–2109. https://doi.org/10.1109/tcyb.2020.2980888 doi: 10.1109/tcyb.2020.2980888
[21]	S. Liu, Q. Lin, L. Feng, K. C. Wong, K. C. Tan, Evolutionary multitasking for large-scale multiobjective optimization, IEEE Trans. Evol. Comput., 2022 (2022), 1–15. https://doi.org/10.1109/tevc.2022.3166482 doi: 10.1109/tevc.2022.3166482
[22]	Y. Cai, D. Peng, S. Fu, H. Tian, Multitasking differential evolution with difference vector sharing mechanism, in 2019 IEEE Symposium Series on Computational Intelligence (SSCI), (2019), 3039–3046. https://doi.org/10.1109/ssci44817.2019.9002698
[23]	X. Chen, Y. Huang, W. Zhou, L. Feng, Evolutionary multitasking via artificial neural networks, in 2021 IEEE International Conference on Systems, Man, and Cybernetics (SMC), (2021), 1545–1552. https://doi.org/10.1109/smc52423.2021.9659031
[24]	S. Huang, J. Zhong, W. J. Yu, Surrogate-assisted evolutionary framework with adaptive knowledge transfer for multi-task optimization, IEEE Trans. Emerging Topics Comput., 9 (2021), 1930–1944. https://doi.org/10.1109/tetc.2019.2945775 doi: 10.1109/tetc.2019.2945775
[25]	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
[26]	X. Ma, Y. Zheng, Z. Zhu, X. Li, L. Wang, Y. Qi, et al., Improving evolutionary multitasking optimization by leveraging inter-task gene similarity and mirror transformation, IEEE Comput. Intell. Mag., 16 (2021), 38–53. https://doi.org/10.1109/mci.2021.3108311 doi: 10.1109/mci.2021.3108311
[27]	A. T. W. Min, Y. S. Ong, A. Gupta, C. K. Goh, Multiproblem surrogates: Transfer evolutionary multiobjective optimization of computationally expensive problems, IEEE Trans. Evol. Comput., 23 (2019), 15–28. https://doi.org/10.1109/tevc.2017.2783441 doi: 10.1109/tevc.2017.2783441
[28]	L. Bai, W. Lin, A. Gupta, Y. S. Ong, From multitask gradient descent to gradient-free evolutionary multitasking: A proof of faster convergence, IEEE Trans. Cybern., 52 (2022), 8561–8573. https://doi.org/10.1109/tcyb.2021.3052509 doi: 10.1109/tcyb.2021.3052509
[29]	A. Gretton, K. M. Borgwardt, M. J. Rasch, A. Smola, B. Schölkopf, A. Smola, A kernel two-sample test, J. Mach. Learn. Res., 13 (2012), 723–773.
[30]	J. Lin, H. L. Liu, B. Xue, M. Zhang, F. Gu, Multiobjective multitasking optimization based on incremental learning, IEEE Trans. Evol. Comput., 24 (2020), 824–838. https://doi.org/10.1109/tevc.2022.3147568 doi: 10.1109/tevc.2022.3147568
[31]	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
[32]	J. Liang, L. Zhang, K. Yu, B. Qu, C. Yue, K. Qiao, A differential evolution based self-adaptive multi-task evolutionary algorithm, in 2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT), (2021), 150–156. https://doi.org/10.1109/acait53529.2021.9731139
[33]	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 2017 IEEE Congress on Evolutionary Computation (CEC), (2017), 921–928. https://doi.org/10.1109/cec.2017.7969407
[34]	H. Chen, H. L. Liu, F. Gu, K. C. Tan, A multi-objective multitask optimization algorithm using transfer rank, IEEE Trans. Evol. Comput., 2022 (2022), 1–15. https://doi.org/10.1109/TEVC.2022.3147568 doi: 10.1109/TEVC.2022.3147568
[35]	J. Lin, H. L. Liu, K. C. Tan, F. Gu, An effective knowledge transfer approach for multiobjective multitasking optimization, IEEE Trans. Cybern., 51 (2021), 3238–3248. https://doi.org/10.1109/tcyb.2020.2969025 doi: 10.1109/tcyb.2020.2969025
[36]	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
[37]	Z. Liang, W. Liang, Z. Wang, X. Ma, L. Liu, Z. Zhu, Multiobjective evolutionary multitasking with two-stage adaptive knowledge transfer based on population distribution, IEEE Trans. Syst. Man Cybern. Syst., 52 (2022), 4457–4469. https://doi.org/10.1109/tsmc.2021.3096220 doi: 10.1109/tsmc.2021.3096220
[38]	W. Zhang, X. Zhang, X. Hao, M. Gen, G. Zhang, W. Yang, Multi-stage hybrid evolutionary algorithm for multiobjective distributed fuzzy flow-shop scheduling problem, Math. Biosci. Eng., 20 (2023), 4838–4864. https://doi.org/10.3934/mbe.2023224 doi: 10.3934/mbe.2023224
[39]	Q. Shang, L. Zhang, L. Feng, Y. Hou, J. Zhong, A. Gupta, et al., A preliminary study of adaptive task selection in explicit evolutionary many-tasking, in 2019 IEEE Congress on Evolutionary Computation (CEC), (2019), 2153–2159. https://doi.org/10.1109/cec.2019.8789909
[40]	Y. Chen, J. Zhong, L. Feng, J. Zhang, An adaptive archive-based evolutionary framework for many-task optimization, IEEE Trans. Emerging Topics Computat. Intell., 4 (2020), 369–384. https://doi.org/10.1109/tetci.2019.2916051 doi: 10.1109/tetci.2019.2916051
[41]	X. Xue, K. Zhang, K. C. Tan, L. Feng, J. Wang, G. Chen, et al., Affine transformation-enhanced multifactorial optimization for heterogeneous problems, IEEE Trans. Cybern., 52 (2022), 6217–6231. https://doi.org/10.1109/tcyb.2020.3036393 doi: 10.1109/tcyb.2020.3036393
[42]	R. Hashimoto, H. Ishibuchi, N. Masuyama, Y. Nojima, Analysis of evolutionary multi-tasking as an island model, in Proceedings of the Genetic and Evolutionary Computation Conference Companion, (2018), 1894–1897. https://doi.org/10.1145/3205651.3208228
[43]	K. K. Bali, A. Gupta, L. Feng, Y. S. Ong, T. P. Siew, 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
[44]	C. Wang, J. Liu, K. Wu, Z. Wu, Solving multi-task optimization problems with adaptive knowledge transfer via anomaly detection, IEEE Trans. Evol. Comput., 26 (2022), 304–318. https://doi.org/10.1109/tevc.2021.3068157 doi: 10.1109/tevc.2021.3068157
[45]	Y. Li, W. Gong, S. Li, Multitasking optimization via an adaptive solver multitasking evolutionary framework, Inf. Sci., 2022 (2022), 1–24. https://doi.org/10.1016/j.ins.2022.10.099 doi: 10.1016/j.ins.2022.10.099
[46]	B. Da, Y. Ong, L. Feng, 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
[47]	R. B. Agrawal, K. Deb, R. B. Agrawal, Simulated binary crossover for continuous search space, Complex Syst., 9 (1994), 115–148.

Reader Comments

Your name:*

Email:*
© 2023 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)