Abstract
Multitasking for multi-objective optimization (MTMO) is one of the most important issues in evolutionary computation. The information exchange mechanism among inter-tasks is the key factor in enhancing the algorithm. Evolutionary multitasking algorithm based on generative strategies (EMT-GS), the current mainstream algorithm, employs a generative adversarial network (GAN) to acquire, propagate and exploit knowledge among tasks, yet GAN suffers from a series of intractable drawbacks, such as training difficulties and mode collapse, etc. To address the issues listed above and achieve better performance, this paper proposes a new algorithm named MTMO-WGAN, which leverages Wasserstein GAN(WGAN) with weight clipping and gradient penalty as the generative strategies to deal with MTMO problems, respectively. Based on the MTMOO benchmark problems, MTMO-WGAN outperforms EMT-GS in the bulk of tasks and has great potential for advancement in the future, which unlocks possibilities for the application of deep generative models in the field of MTMO.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Back, T., Hammel, U., Schwefel, H.-P.: Evolutionary computation: comments on the history and current state. IEEE Trans. Evol. Comput. 1, 3–17 (1997)
Tanabe, R., Ishibuchi, H.: A review of evolutionary multimodal multiobjective optimization. IEEE Trans. Evol. Comput. 24, 193–200 (2019)
Gupta, A., Ong, Y.-S., Feng, L.: Multifactorial evolution: toward evolutionary multitasking. IEEE Trans. Evol. Comput. 20, 343–357 (2015)
Gupta, A., Ong, Y.-S., Feng, L., Tan, K.C.: Multiobjective multifactorial optimization in evolutionary multitasking. IEEE Trans. Cybern. 47, 1652–1665 (2016)
Wang X., Dong Z., Tang L., Zhang Q.: Multiobjective multitask optimization-neighborhood as a bridge for knowledge transfer. IEEE Trans. Evol. Comput. (2022)
Chen, Z., Zhou, Y., He, X., Zhang, J.: Learning task relationships in evolutionary multitasking for multiobjective continuous optimization. IEEE Trans. Cybern. 52(6), 5278–5289 (2020)
Lin, J., Liu, H.-L., Xue, B., Zhang, M., Gu, F.: Multiobjective multitasking optimization based on incremental learning. IEEE Trans. Evol. Comput. 24, 824–838 (2019)
Liang, Z., Zhang, J., Feng, L., Zhu, Z.: A hybrid of genetic transform and hyper-rectangle search strategies for evolutionary multi-tasking. Expert Syst. Appl. 138, 112798 (2019)
Liang, Z., Dong, H., Liu, C., Liang, W., Zhu, Z.: Evolutionary multitasking for multiobjective optimization with subspace alignment and adaptive differential evolution. IEEE Trans. Cybern. 52(4), 2096–2109 (2020)
Lin, J., Liu, H.-L., Tan, K.C., Gu, F.: An effective knowledge transfer approach for multiobjective multitasking optimization. IEEE Trans. Cybern. 51, 3238–3248 (2020)
Liang, Z., Liang, W., Wang, Z., Ma, X., Liu, L., Zhu, Z.: Multiobjective evolutionary multitasking with two-stage adaptive knowledge transfer based on population distribution. IEEE Trans. Syst. Man Cybern. Syst. 52(7), 4457–4469 (2021)
Bali, K.K., Gupta, A., Ong, Y.-S., Tan, P.S.: Cognizant multitasking in multiobjective multifactorial evolution: mo-mfea-Ii. IEEE Trans. Cybern. 51, 1784–1796 (2020)
Yang, C., Ding, J., Tan, K.C., Jin, Y.: Two-stage assortative mating for multi-objective multifactorial evolutionary optimization. In: 2017 IEEE 56th Annual Conference on Decision and Control (CDC), pp. 76–81. IEEE (2017)
Xu, Z., Liu, X., Zhang, K., He, J.: Cultural transmission based multi-objective evolution strategy for evolutionary multitasking. Inf. Sci. 582, 215–242 (2022)
Binh, H.T.T., Tuan, N.Q., Long, D.C.T.: A multi-objective multi-factorial evolutionary algorithm with reference-point-based approach. In: 2019 IEEE Congress on Evolutionary Computation (CEC), pp. 2824–2831. IEEE (2019)
Yao, S., Dong, Z., Wang, X., Ren, L.: A multiobjective multifactorial optimization algorithm based on decomposition and dynamic resource allocation strategy. Inf. Sci. 511, 18–35 (2020)
Hashimoto, R., Urita, T., Masuyama, N., Nojima, Y., Ishibuchi, H.: Effects of local mating in inter-task crossover on the performance of decomposition-based evolutionary multiobjective multitask optimization algorithms. In: 2020 IEEE Congress on Evolutionary Computation (CEC), pp. 1–8. IEEE (2020)
Wei, T., Zhong, J.: Towards generalized resource allocation on evolutionary multitasking for multi-objective optimization. IEEE Comput. Intell. Mag. 16, 20–37 (2021)
Chen, Y., Zhong, J., Tan, M.: A fast memetic multi-objective differential evolution for multi-tasking optimization. In: 2018 IEEE Congress on Evolutionary Computation (CEC), pp. 1–8. IEEE (2018)
Zhou, L., et al.: Toward adaptive knowledge transfer in multifactorial evolutionary computation. IEEE Trans. Cybern. 51, 2563–2576 (2020)
Liang, Z., Zhu, Y., Wang, X., Li, Z., Zhu, Z.: Evolutionary multitasking for multi-objective optimization based on generative strategies. IEEE Trans. Evol. Comput. 52(4), 2096–2109 (2022)
Feng, L., et al.: Evolutionary multitasking via explicit autoencoding. IEEE Trans. Cybern. 49, 3457–3470 (2018)
Chen, H., Liu, H.-L., Gu, F., Tan, K.C.: A multi-objective multitask optimization algorithm using transfer rank. IEEE Trans. Evol. Comput. (2022)
Goodfellow, I., et al.: Generative adversarial networks. Commun. ACM 63, 139–144 (2020)
Arjovsky, M., Chintala, S., Bottou, L.: Wasserstein generative adversarial networks. In: International Conference on Machine Learning, pp. 214–223. PMLR (2017)
Villani, C.: Optimal Transport: Old and New. Springer, Heidelberg (2009). https://doi.org/10.1007/978-3-540-71050-9
Gulrajani, I., Ahmed, F., Arjovsky, M., Dumoulin, V., Courville, A.C.: Improved training of wasserstein GANs. In: Advances in Neural Information Processing Systems, vol. 30 (2017)
Zitzler, E., Thiele, L., Laumanns, M., Fonseca, C.M., Da Fonseca, V.G.: Performance assessment of multiobjective optimizers: an analysis and review. IEEE Trans. Evol. Comput. 7, 117–132 (2003)
Yuan, Y., et al.: Evolutionary Multitasking for Multiobjective Continuous Optimization: Benchmark Problems, Performance Metrics and Baseline Results. arXiv preprint arXiv:1706.02766 (2017)
Acknowledgment
The study was supported in part by the Natural Science Foundation of China under Grant No. 62103286, No. 62001302, in part by Ministry of Education of the People’s Republic of China Humanities and Social Sciences Youth Foundation under Grant No. 21YJC630181, in part by Guangdong Basic and Applied Basic Research Foundation under Grant No. 2021A1515011348, in part by Natural Science Foundation of Shenzhen under Grant No. JCYJ20190808145011259, in part by Natural Science Foundation of Guangdong Province under Grant No. 2020A1515010752, No. 2020A1515110541, in part by Guangdong Province Innovation Team under Grant 2021WCXTD002, in part by Shenzhen Science and Technology Program under Grant RCBS20200714114920379.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Zhou, T., Yao, X., Yue, G., Niu, B. (2023). A WGAN-Based Generative Strategy in Evolutionary Multitasking for Multi-objective Optimization. In: Tan, Y., Shi, Y., Luo, W. (eds) Advances in Swarm Intelligence. ICSI 2023. Lecture Notes in Computer Science, vol 13968. Springer, Cham. https://doi.org/10.1007/978-3-031-36622-2_32
Download citation
DOI: https://doi.org/10.1007/978-3-031-36622-2_32
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-36621-5
Online ISBN: 978-3-031-36622-2
eBook Packages: Computer ScienceComputer Science (R0)