Abstract
The parameter setting problem constitutes one of the major challenges in evolutionary computation, and is subject to considerable research efforts. Since the optimal parameter values can change during the optimization process, efficient parameter control techniques that automatically identify and track reasonable parameter values are sought.
A potential drawback of dynamic parameter selection is that state-of-the-art control mechanisms introduces themselves new sets of hyper-parameters, which need to be tuned for the problem at hand. The general hope is that the performance of an algorithm is much less sensitive with respect to these hyper-parameters than with respect to its original parameters. This belief is backed up by a number of empirical and theoretical results. What is less understood in discrete black-box optimization, however, is the influence of the initial parameter value. We contribute with this work an empirical sensitivity analysis for three selected algorithms with self-adjusting parameter choices: the (1 + 1) EA\(_{\alpha }\), the 2-rate \((1+\lambda )\) EA\(_{2r,r/2}\), and the \((1+(\lambda ,\lambda ))\) GA. In all three cases we observe fast convergence of the parameters towards their optimal choices. The performance loss of a sub-optimal initialization is shown to be almost negligible for the former two algorithms. For the \((1+(\lambda ,\lambda ))\) GA, in contrast, the choice of \(\lambda \) is more critical; our results suggest to initialize it by a small value.
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 subscriptionsNotes
- 1.
We remark that a one-way ANOVA is not applicable as the Shapiro-Wilk normality test returns that the data is not normally distributed.
References
Aleti, A., Moser, I.: A systematic literature review of adaptive parameter control methods for evolutionary algorithms. ACM Comput. Surv. 49, 56:1–56:35 (2016)
Ansótegui, C., Malitsky, Y., Samulowitz, H., Sellmann, M., Tierney, K.: Model-based genetic algorithms for algorithm configuration. In: IJCAI 2015, pp. 733–739. AAAI Press (2015)
Bartz-Beielstein, T.: SPOT: an R package for automatic and interactive tuning of optimization algorithms by sequential parameter optimization. CoRR abs/1006.4645 (2010). http://arxiv.org/abs/1006.4645
Böttcher, S., Doerr, B., Neumann, F.: Optimal fixed and adaptive mutation rates for the leadingones problem. In: Schaefer, R., Cotta, C., Kołodziej, J., Rudolph, G. (eds.) PPSN 2010. LNCS, vol. 6238, pp. 1–10. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-15844-5_1
Devroye, L.: The compound random search. Ph.D. dissertation, Purdue University, West Lafayette, IN (1972)
Doerr, B., Doerr, C.: Optimal static and self-adjusting parameter choices for the \((1+(\lambda,\lambda ))\) genetic algorithm. Algorithmica 80, 1658–1709 (2018)
Doerr, B., Doerr, C.: Theory of parameter control mechanisms for discrete black-box optimization: provable performance gains through dynamic parameter choices. In: Doerr, B., Neumann, F. (eds.) Theory of Randomized Search Heuristics in Discrete Search Spaces. Springer, Cham (2018, to appear). https://arxiv.org/abs/1804.05650
Doerr, B., Doerr, C., Ebel, F.: From black-box complexity to designing new genetic algorithms. Theor. Comput. Sci. 567, 87–104 (2015)
Doerr, B., Doerr, C., Yang, J.: Optimal parameter choices via precise black-box analysis. In: GECCO 2016, pp. 1123–1130. ACM (2016)
Doerr, B., Gießen, C., Witt, C., Yang, J.: The \((1+\lambda )\) evolutionary algorithm with self-adjusting mutation rate. In: GECCO 2017, pp. 1351–1358. ACM (2017)
Doerr, C., Wagner, M.: On the effectiveness of simple success-based parameter selection mechanisms for two classical discrete black-box optimization benchmark problems. In: GECCO 2018. ACM (2018, to appear). https://arxiv.org/abs/1803.01425
Eiben, A.E., Hinterding, R., Michalewicz, Z.: Parameter control in evolutionary algorithms. IEEE Trans. Evol. Comput. 3, 124–141 (1999)
Hansen, N., Ostermeier, A.: Completely derandomized self-adaptation in evolution strategies. Evol. Comput. 9, 159–195 (2001)
Hutter, F., Hoos, H.H., Leyton-Brown, K.: Sequential model-based optimization for general algorithm configuration. In: Coello, C.A.C. (ed.) LION 2011. LNCS, vol. 6683, pp. 507–523. Springer, Heidelberg (2011). https://doi.org/10.1007/978-3-642-25566-3_40
Hutter, F., Hoos, H.H., Leyton-Brown, K., Stützle, T.: ParamILS: an automatic algorithm configuration framework. J. Artif. Intell. Res. 36, 267–306 (2009)
Karafotias, G., Hoogendoorn, M., Eiben, A.: Parameter control in evolutionary algorithms: trends and challenges. IEEE Trans. Evol. Comput. 19, 167–187 (2015)
Lobo, F.J., Lima, C.F., Michalewicz, Z. (eds.): Parameter Setting in Evolutionary Algorithms. Studies in Computational Intelligence, vol. 54. Springer, Heidelberg (2007). https://doi.org/10.1007/978-3-540-69432-8
López-Ibáñez, M., Dubois-Lacoste, J., Cáceres, L.P., Birattari, M., Stützle, T.: The irace package: iterated racing for automatic algorithm configuration. Oper. Res. Perspect. 3, 43–58 (2016)
Rechenberg, I.: Evolutionsstrategie. Friedrich Fromman Verlag (Günther Holzboog KG), Stuttgart (1973)
Schumer, M.A., Steiglitz, K.: Adaptive step size random search. IEEE Trans. Autom. Control 13, 270–276 (1968)
Thierens, D.: On benchmark properties for adaptive operator selection. In: Companion Material GECCO 2009, pp. 2217–2218. ACM (2009)
Acknowledgments
We would like to thank Eduardo Carvalho Pinto and Christian Giessen for providing their implementations of the \((1 + 1)\) EA\(_{\alpha }\) and the \((1 + (\lambda ,\lambda ))\) GA and the \((1+\lambda )\) EA\(_{r/2,2r}\), respectively.
Our work was supported by a public grant as part of the Investissement d’avenir project, reference ANR-11-LABX-0056-LMH, LabEx LMH and by the Australian Research Council project DE160100850.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Doerr, C., Wagner, M. (2018). Sensitivity of Parameter Control Mechanisms with Respect to Their Initialization. In: Auger, A., Fonseca, C., Lourenço, N., Machado, P., Paquete, L., Whitley, D. (eds) Parallel Problem Solving from Nature – PPSN XV. PPSN 2018. Lecture Notes in Computer Science(), vol 11102. Springer, Cham. https://doi.org/10.1007/978-3-319-99259-4_29
Download citation
DOI: https://doi.org/10.1007/978-3-319-99259-4_29
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-99258-7
Online ISBN: 978-3-319-99259-4
eBook Packages: Computer ScienceComputer Science (R0)