Effect of transformations of numerical parameters in automatic algorithm configuration

  • Alberto Franzin
  • Leslie Pérez Cáceres
  • Thomas Stützle
Original Paper
  • 32 Downloads

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.

Keywords

Automatic algorithm configuration Numerical parameters Parameter tuning Parameter transformation 

Notes

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.

References

  1. 1.
    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)Google Scholar
  2. 2.
    Çela, E.: The Quadratic Assignment Problem: Theory and Algorithms. Kluwer Academic Publishers, Dordrecht (1998)CrossRefMATHGoogle Scholar
  3. 3.
    Cohn, H., Fielding, M.J.: Simulated annealing: searching for an optimal temperature. SIAM J. Optim. 9(3), 779–802 (1999)MathSciNetCrossRefMATHGoogle Scholar
  4. 4.
    Eiben, A.E., Smit, S.K.: Parameter tuning for configuring and analyzing evolutionary algorithms. Swarm Evol. Comput. 1(1), 19–31 (2011)CrossRefGoogle Scholar
  5. 5.
    Fielding, M.J.: Simulated annealing with an optimal fixed temperature. SIAM J. Optim. 11(2), 289–307 (2000)MathSciNetCrossRefMATHGoogle Scholar
  6. 6.
    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)CrossRefGoogle Scholar
  7. 7.
    Hoos, H.H.: Programming by optimization. Commun. ACM 55(2), 70–80 (2012)CrossRefGoogle Scholar
  8. 8.
    Hussin, M.S., Stützle, T.: Tabu search vs. simulated annealing for solving large quadratic assignment instances. Comput. Oper. Res. 43, 286–291 (2014)CrossRefMATHGoogle Scholar
  9. 9.
    Hutter, F., Hoos, H.H., Leyton-Brown, K., Stützle, T.: ParamILS: an automatic algorithm configuration framework. J. Artif. Intell. Res. 36, 267–306 (2009)MATHGoogle Scholar
  10. 10.
    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)Google Scholar
  11. 11.
    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)CrossRefGoogle Scholar
  12. 12.
    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)MathSciNetCrossRefGoogle Scholar
  13. 13.
    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)CrossRefGoogle Scholar
  14. 14.
    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)Google Scholar
  15. 15.
    Taillard, É.D.: Robust taboo search for the quadratic assignment problem. Parallel Comput. 17(4–5), 443–455 (1991)MathSciNetCrossRefGoogle Scholar
  16. 16.
    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)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  1. 1.IRIDIAUniversité Libre de Bruxelles (ULB)BrusselsBelgium

Personalised recommendations