Abstract
Data-driven evolutionary algorithms (DDEAs) have attracted much attention in recent years, due to their effectiveness and advantages in solving expensive and complex optimization problems. In an offline data-driven evolutionary algorithm, no additional information is available to update surrogate models, so we need to ensure the quality of the surrogate models built before optimization. Based on the offline data-driven evolutionary algorithm using selective surrogate ensembles (DDEA-SE), we propose a trimmed bagging based DDEA-SE (TDDEA-SE) to construct a more accurate model pool. To this end, we use trimmed bagging to prune models with large errors calculated by out-of-bag samples, thus improving the accuracy of the selective surrogate ensembles and promoting the optimization. The experimental results on benchmark problems show that the proposed algorithm can strike a balance between diversity and accuracy of models, and its competitiveness on solving offline data-driven optimization problems.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Jin, Y.C.: Surrogate-assisted evolutionary computation: recent advances and future challenges. Swarm Evol. Comput. 1(2), 61–70 (2011)
Vikhar, P.A.: Evolutionary algorithms: A critical review and its future prospects. In: International Conference on Global Trends in Signal Processing, Information Computing and Communication, ICGTSPICC 2016, 22–24 December, Jalgaon, India, pp. 261–265 (2016)
Pan, L.Q., Li, L.H., He, C., et al.: A subregion division-based evolutionary algorithm with effective mating selection for many-objective optimization. IEEE Trans. Cybern. 50(8), 3477–3490 (2020)
Pan, L.Q., Li, L.H., He, C., et al.: Manifold learning-inspired mating restriction for evolutionary multiobjective optimization with complicated pareto sets. IEEE Trans. Cybern. 1–13 (2019). https://doi.org/10.1109/TCYB.2019.2952881
Wang, H.D., Jin, Y.C., Doherty, J.: Committee-based active learning for surrogate-assisted particle swarm optimization of expensive problems. IEEE Trans. Cybern. 47(9), 2664–2677 (2017)
Nguyen, S., Zhang, M.J., Johnston, M., et al.: Automatic programming via iterated local search for dynamic job shop scheduling. IEEE Trans. Cybern. 45(1), 1–14 (2015)
Wang, S., Liu, J., Jin, Y.C.: Surrogate-assisted robust optimization of large-scale networks based on graph embedding. IEEE Trans. Evol. Comput. 24(4), 735–749 (2020)
Guo, D., Chai, T.Y., Ding, J., et al.: Small data driven evolutionary multi-objective optimization of fused magnesium furnaces. In: IEEE Symposium Series on Computational Intelligence, IEEE SSCI 2016, 6–9 December, Athens, Greece, pp. 1–8 (2016)
Lian, Y.S., Liou, M.S.: Multiobiective optimization using coupled response surface model and evolutionary algorithm. AIAA J. 43(6), 1316–1325 (2005)
Liu, B., Zhang, Q.F., Gielen, G., et al.: A gaussian process surrogate model assisted evolutionary algorithm for medium scale expensive optimization problems. IEEE Trans. Evol. Comput. 18(2), 180–192 (2014)
Wang, H.D., Jin, Y.C., Sun, C.L., et al.: Offline data-driven evolutionary optimization using selective surrogate ensembles. IEEE Trans. Evol. Comput. 23(2), 203–216 (2019)
Sun, C.L., Jin, Y.C., Cheng, R., et al.: Surrogate-assisted cooperative swarm optimization of high-dimensional expensive problems. IEEE Trans. Evol. Comput. 21(4), 644–660 (2017)
Pan, L.Q., He, C., Tian, Y., et al.: A classification-based surrogate-assisted evolutionary algorithm for expensive many-objective optimization. IEEE Trans. Evol. Comput. 23(1), 74–88 (2019)
Jin, Y.C., Olhofer, M., Sendhoff, B.: A framework for evolutionary optimization with approximate fitness functions. IEEE Trans. Evol. Comput. 6(5), 481–494 (2002)
Li, J.Y., Zhan, Z.H., Wang, C., et al.: Boosting data-driven evolutionary algorithm with localized data generation. IEEE Trans. Evol. Comput. (2020, to be published). https://doi.org/10.1109/TEVC.2020.2979740
Yang, C.E., Ding, J.L., Jin, Y.C., et al.: Offline data-driven multiobjective optimization: knowledge transfer between surrogates and generation of final solutions. IEEE Trans. Evol. Comput. 24(3), 409–423 (2020)
Croux, C., Jooseens, K., Lemmensb, A.: Trimmed bagging. Comput. Stat. Data Anal. 52(1), 362–368 (2007)
Zhang, H., Song, Y.J., Jiang, B., et al.: Two-stage bagging pruning for reducing the ensemble size and improving the classification performance. Math. Prob. Eng. 5(3), 1–17 (2019)
Moreira, J.M., Soares, C., Jorge, A.M., et al.: Ensemble approaches for regression: a survey. ACM Comput. Surv. 45(1), 1–40 (2012)
Zhang, C.X., Guo, G.: Research of the applications of out-of-bag sample. Software 32(3), 1–4 (2011)
Bylander, T.: Estimating generalization error on two class datasets using out-of-bag estimates. Mach. Learn. 48(1), 287–297 (2002)
Rao, J.S., Tibshirant, R.: The out-of-bootstrap method for model averaging and selection. Technical report, Cleveland Clinic and University of Toronto (997)
Derrac, J., Garcia, S., Herrera, F.: A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms. Swarm Evol. Comput. 1(1), 3–18 (2011)
Acknowledgements
This work is supported by the National Key R and D Program of China for International S and T Cooperation Projects (2017YFE0103900), China Postdoctoral Science Foundation (2020M672359) and the Fundamental Research Funds for the Central Universities (HUST: 2019kfyXMBZ056).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Shan, Y., Hou, Y., Wang, M., Xu, F. (2021). Trimmed Data-Driven Evolutionary Optimization Using Selective Surrogate Ensembles. In: Pan, L., Pang, S., Song, T., Gong, F. (eds) Bio-Inspired Computing: Theories and Applications. BIC-TA 2020. Communications in Computer and Information Science, vol 1363. Springer, Singapore. https://doi.org/10.1007/978-981-16-1354-8_10
Download citation
DOI: https://doi.org/10.1007/978-981-16-1354-8_10
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-1353-1
Online ISBN: 978-981-16-1354-8
eBook Packages: Computer ScienceComputer Science (R0)