Abstract
Well-defined Hyper-Heuristics enhance the generalization of MOEAs and the blind usability on complex and even dynamic real-world application. Previous works already showed, that Hyper-Heuristics as selectors of crossover operators improve the performance of a single algorithm used on two opposing problem properties. In this paper, we present different selection mechanisms of Hyper-Heuristics, that are able to handle an expanded selection pool to cover more properties. We solve 20 benchmark problems with NSGA-II using those Hyper-Heuristics. By comparing the learning behaviour and the IGD trends of fixed crossover operator usages, we confirm that a combination of operators could outperform the best fixed operator. From the introduced Hyper-Heuristics in this paper, HHX-A made the best use of this advantage. It selects either all operators or a single operator alternately and learns fast which operators to prioritize to optimize the production. Due to periodic resets of the score, the Hyper-Heuristic is able to adapt fast to changes of the current state of the solving process. Although the pool is bigger and more diverse, we are able to show that HHX-A decides reasonably and fast. Therefore, it works well on a bigger set of problems with different properties.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Beyer, H.G., Deb, K.: On self-adaptive features in real-parameter evolutionary algorithms. IEEE Trans. Evol. Comput. 5(3), 250–270 (2001). https://doi.org/10.1109/4235.930314
Burke, E.K., et al.: Hyper-heuristics: a survey of the state of the art. J. Oper. Res. Soc. 64(12), 1695–1724 (2013). https://doi.org/10.1057/jors.2013.71
Coello Coello, C.A., Reyes Sierra, M.: A study of the parallelization of a coevolutionary multi-objective evolutionary algorithm. In: Monroy, R., Arroyo-Figueroa, G., Sucar, L.E., Sossa, H. (eds.) MICAI 2004. LNCS (LNAI), vol. 2972, pp. 688–697. Springer, Heidelberg (2004). https://doi.org/10.1007/978-3-540-24694-7_71
Deb, K., Joshi, D., Anand, A.: Real-coded evolutionary algorithms with parent-centric recombination. In: Proceedings of the 2002 Congress on Evolutionary Computation, CEC 2002 (Cat. No.02TH8600), vol. 1, pp. 61–66 (2002). https://doi.org/10.1109/CEC.2002.1006210
Deb, K., Pratap, A., Agarwal, S., Meyarivan, T.: A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans. Evol. Comput. 6(2), 182–197 (2002). https://doi.org/10.1109/4235.996017
Deb, K., Agrawal, R.B.: Simulated binary crossover for continuous search space. Complex Syst. 9, 115–148 (1995)
Deb, K., Thiele, L., Laumanns, M., Zitzler, E.: Scalable test problems for evolutionary multiobjective optimization. In: Abraham, A., Jain, L., Goldberg, R. (eds.) Evolutionary Multiobjective Optimization. Advanced Information and Knowledge Processing, pp. 105–145. Springer, London (2005). https://doi.org/10.1007/1-84628-137-7_6
Deep, K., Thakur, M.: A new crossover operator for real coded genetic algorithms. Appl. Math. Comput. 188(1), 895–911 (2007). https://doi.org/10.1016/j.amc.2006.10.047, https://linkinghub.elsevier.com/retrieve/pii/S0096300306014287
Drake, J., Kheiri, A., Özcan, E., Burke, E.: Recent advances in selection hyper-heuristics. Eur. J. Oper. Res. 285, 405–428 (2020). https://doi.org/10.1016/j.ejor.2019.07.073
Fritsche, G., Pozo, A.: A hyper-heuristic collaborative multi-objective evolutionary algorithm. In: 2018 7th Brazilian Conference on Intelligent Systems (BRACIS), pp. 354–359 (2018). https://doi.org/10.1109/BRACIS.2018.00068
Hansen, N., Ostermeier, A.: Completely derandomized self-adaptation in evolution strategies. Evol. Comput. 9(2), 159–195 (2001). https://doi.org/10.1162/106365601750190398
Hong, L., Drake, J.H., Woodward, J.R., Özcan, E.: Automatically designing more general mutation operators of evolutionary programming for groups of function classes using a hyper-heuristic. In: Proceedings of the Genetic and Evolutionary Computation Conference 2016, pp. 725–732. ACM, Denver Colorado (2016). https://doi.org/10.1145/2908812.2908958, https://dl.acm.org/doi/10.1145/2908812.2908958
Huband, S., Barone, L., While, L., Hingston, P.: A scalable multi-objective test problem toolkit. In: Coello Coello, C.A., Hernández Aguirre, A., Zitzler, E. (eds.) EMO 2005. LNCS, vol. 3410, pp. 280–295. Springer, Heidelberg (2005). https://doi.org/10.1007/978-3-540-31880-4_20
Ono, I., Kita, H., Kobayashi, S.: A robust real-coded genetic algorithm using unimodal normal distribution crossover augmented by uniform crossover: effects of self-adaptation of crossover probabilities, p. 8
Pan, L., Xu, W., Li, L., He, C., Cheng, R.: Adaptive simulated binary crossover for rotated multi-objective optimization. Swarm Evol. Comput. 60, 100759 (2021). https://doi.org/10.1016/j.swevo.2020.100759, https://www.sciencedirect.com/science/article/pii/S2210650220304120
Pang, L.M., Ishibuchi, H., Shang, K.: Using a genetic algorithm-based hyper-heuristic to tune MOEA/D for a set of various test problems. In: 2021 IEEE Congress on Evolutionary Computation (CEC), pp. 1486–1494 (2021). https://doi.org/10.1109/CEC45853.2021.9504748
Pillay, N., Qu, R.: Hyper-Heuristics: Theory and Applications. Natural Computing Series, Springer, Cham (2018). https://doi.org/10.1007/978-3-319-96514-7
Qingfu Zhang, Aimin Zhou, Yaochu Jin: RM-MEDA: a regularity model-based multiobjective estimation of distribution algorithm. IEEE Trans. Evol. Compu. 12(1), 41–63 (2008). https://doi.org/10.1109/TEVC.2007.894202, http://ieeexplore.ieee.org/document/4358761/
Ross, P.: Hyper-heuristics. In: Burke, E.K., Kendall, G. (eds.) Search Methodologies: Introductory Tutorials in Optimization and Decision Support Techniques, pp. 529–556. Springer, Boston (2005). https://doi.org/10.1007/0-387-28356-0_17
Storn, R.: Differential evolution - a simple and efficient adaptive scheme for global optimization over continuous spaces, p. 16
Syswerda, G.: Uniform crossover in genetic algorithms (1989)
Tian, Y., Cheng, R., Zhang, X., Jin, Y.: PlatEMO: a MATLAB platform for evolutionary multi-objective optimization. IEEE Comput. Intell. Mag. 12(4), 73–87 (2017)
Tsutsui, S., Yamamura, M., Higuchi, T.: Multi-parent recombination with simplex crossover in real coded genetic algorithms, p. 9
Venske, S.M., Almeida, C.P., Delgado, M.R.: Comparing selection hyper-heuristics for many-objective numerical optimization. In: 2021 IEEE Congress on Evolutionary Computation (CEC), pp. 1921–1928 (2021). https://doi.org/10.1109/CEC45853.2021.9504934
Zitzler, E., Deb, K., Thiele, L.: Comparison of multiobjective evolutionary algorithms: empirical results. Evol. Comput. 8(2), 173–195 (2000). https://doi.org/10.1162/106365600568202, https://direct.mit.edu/evco/article/8/2/173-195/868
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
Heise, J., Mostaghim, S. (2023). Online Learning Hyper-Heuristics in Multi-Objective Evolutionary Algorithms. In: Emmerich, M., et al. Evolutionary Multi-Criterion Optimization. EMO 2023. Lecture Notes in Computer Science, vol 13970. Springer, Cham. https://doi.org/10.1007/978-3-031-27250-9_12
Download citation
DOI: https://doi.org/10.1007/978-3-031-27250-9_12
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-27249-3
Online ISBN: 978-3-031-27250-9
eBook Packages: Computer ScienceComputer Science (R0)