Abstract
We study the impact of altering the sampling space of parameters in automatic algorithm configurators. We show that a proper transformation can strongly improve the convergence towards better configurations; at the same time, biases about good parameter values, possibly based on misleading prior knowledge, may lead to wrong choices in the transformations and be detrimental for the configuration process. To emphasize the impact of the transformations, we initially study their effect on configuration tasks with a single parameter in different experimental settings. We also propose a mechanism for how to adapt towards an appropriate transformation and give exemplary experimental results of that scheme.
Similar content being viewed by others
Notes
Note that SMAC offers the possibility of a \(\mathcal {L}og(x)\) transformation but not that of \(\mathcal {RL}og(x)\).
References
Ansótegui, C., Sellmann, M., Tierney, K.: A gender-based genetic algorithm for the automatic configuration of algorithms. In: Gent, I.P. (ed.) Principles and Practice of Constraint Programming. CP 2009, volume 5732 of LNCS, pp. 142–157. Springer, Heidelberg (2009)
Çela, E.: The Quadratic Assignment Problem: Theory and Algorithms. Kluwer Academic Publishers, Dordrecht (1998)
Cohn, H., Fielding, M.J.: Simulated annealing: searching for an optimal temperature. SIAM J. Optim. 9(3), 779–802 (1999)
Eiben, A.E., Smit, S.K.: Parameter tuning for configuring and analyzing evolutionary algorithms. Swarm Evol. Comput. 1(1), 19–31 (2011)
Fielding, M.J.: Simulated annealing with an optimal fixed temperature. SIAM J. Optim. 11(2), 289–307 (2000)
Girerd, N., Rabilloud, M., Pibarot, P., Mathieu, P., Roy, P.: Quantification of treatment effect modification on both an additive and multiplicative scale. PLoS ONE 11(4), 1–14 (2016)
Hoos, H.H.: Programming by optimization. Commun. ACM 55(2), 70–80 (2012)
Hussin, M.S., Stützle, T.: Tabu search vs. simulated annealing for solving large quadratic assignment instances. Comput. Oper. Res. 43, 286–291 (2014)
Hutter, F., Hoos, H.H., Leyton-Brown, K., Stützle, T.: ParamILS: an automatic algorithm configuration framework. J. Artif. Intell. Res. 36, 267–306 (2009)
Hutter, F., Hoos, H.H., Leyton-Brown, K.: Sequential model-based optimization for general algorithm configuration. In: Coello Coello, C.A. (ed.) Learning and Intelligent Optimization, 5th International Conference, LION 5, volume 6683 of LNCS, pp. 507–523. Springer, Heidelberg (2011)
Knol, M.J., VanderWeele, T.J., Groenwold, R.H.H., Klungel, O.H., Rovers, M.M., Grobbee, D.E.: Estimating measures of interaction on an additive scale for preventive exposures. Eur. J. Epidemiol. 26(6), 433–438 (2011)
López-Ibáñez, M., Dubois-Lacoste, J., Pérez Cáceres, L., Stützle, T., Birattari, M.: The irace package: iterated racing for automatic algorithm configuration. Oper. Res. Perspect. 3, 43–58 (2016)
Metropolis, N., Rosenbluth, A.W., Rosenbluth, M.N., Teller, A., Teller, E.: Equation of state calculations by fast computing machines. J. Chem. Phys. 21, 1087–1092 (1953)
Snoek, J., Swersky, K., Zemel, R., Adams, R.P.: Input warping for Bayesian optimization of non-stationary functions. In: Proceedings of the 31th International Conference on Machine Learning, vol. 32, pp. 1674–1682 (2014)
Taillard, É.D.: Robust taboo search for the quadratic assignment problem. Parallel Comput. 17(4–5), 443–455 (1991)
Yuan, Z., Montes de Oca, M.A., Stützle, T., Birattari, M.: Continuous optimization algorithms for tuning real and integer algorithm parameters of swarm intelligence algorithms. Swarm Intell. 6(1), 49–75 (2012)
Acknowledgements
This research has received funding from the COMEX Project (P7/36) within the Interuniversity Attraction Poles Programme of the Belgian Science Policy Office. Thomas Stützle acknowledges support from the Belgian F.R.S.-FNRS, of which he is a Research Director.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Franzin, A., Pérez Cáceres, L. & Stützle, T. Effect of transformations of numerical parameters in automatic algorithm configuration. Optim Lett 12, 1741–1753 (2018). https://doi.org/10.1007/s11590-018-1240-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11590-018-1240-3