Advertisement

On the Design of a Master-Worker Adaptive Algorithm Selection Framework

  • Christopher Jankee
  • Sébastien Verel
  • Bilel Derbel
  • Cyril Fonlupt
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10764)

Abstract

We investigate the design of a master-worker schemes for adaptive algorithm selection with the following two-fold goal: (i) choose accurately from a given portfolio a set of operators to be executed in parallel, and consequently (ii) take full advantage of the compute power offered by the underlying distributed environment. In fact, it is still an open issue to design online distributed strategies that are able to optimally assign operators to parallel compute resources when distributively solving a given optimization problem. In our proposed framework, we adopt a reward-based perspective and investigate at what extent the average or maximum rewards collected at the master from the workers are appropriate. Moreover, we investigate the design of both homogeneous and heterogeneous scheme. Our comprehensive experimental study, conducted through a simulation-based methodology and using a recently proposed benchmark family for adaptive algorithm selection, reveals the accuracy of the proposed framework while providing new insights on the performance of distributed adaptive optimization algorithms.

References

  1. 1.
    Armas, R., Aguirre, H., Zapotecas-Martínez, S., Tanaka, K.: Traffic signal optimization: minimizing travel time and fuel consumption. In: Bonnevay, S., Legrand, P., Monmarché, N., Lutton, E., Schoenauer, M. (eds.) EA 2015. LNCS, vol. 9554, pp. 29–43. Springer, Cham (2016).  https://doi.org/10.1007/978-3-319-31471-6_3 Google Scholar
  2. 2.
    Auer, P., Cesa-Bianchi, N., Fischer, P.: Finite-time analysis of the multiarmed bandit problem. Mach. Learn. 47(2–3), 235–256 (2002)CrossRefzbMATHGoogle Scholar
  3. 3.
    Baudiš, P., Pošík, P.: Online black-box algorithm portfolios for continuous optimization. In: Bartz-Beielstein, T., Branke, J., Filipič, B., Smith, J. (eds.) PPSN 2014. LNCS, vol. 8672, pp. 40–49. Springer, Cham (2014).  https://doi.org/10.1007/978-3-319-10762-2_4 Google Scholar
  4. 4.
    Candan, C., Goëffon, A., Lardeux, F., Saubion, F.: Non stationary operator selection with island models. In: GECCO, pp. 1509–1516 (2013)Google Scholar
  5. 5.
    DaCosta, L., Fialho, A., Schoenauer, M., Sebag, M.: Adaptive operator selection with dynamic multi-armed bandits. In: GECCO, p. 913. ACM Press (2008)Google Scholar
  6. 6.
    Dasgupta, D., Michalewicz, Z.: Evolutionary Algorithms in Engineering Applications. Springer, Heidelberg (2013)zbMATHGoogle Scholar
  7. 7.
    Derbel, B., Verel, S.: DAMS: distributed adaptive metaheuristic selection. In: GECCO, pp. 1955–1962. ACM Press (2011)Google Scholar
  8. 8.
    Dubreuil, M., Gagne, C., Parizeau, M.: Analysis of a master-slave architecture for distributed evolutionary computations. IEEE Trans. Syst. Man Cybern. Part B 36, 229–235 (2006)CrossRefzbMATHGoogle Scholar
  9. 9.
    Eiben, A.E., Michalewicz, Z., Schoenauer, M., Smith, J.E.: Parameter control in evolutionary algorithms. In: Parameter Setting in Evolutionary Algorithms, pp. 19–46. Springer, Heidelberg (2007)Google Scholar
  10. 10.
    Fialho, Á., Da Costa, L., Schoenauer, M., Sebag, M.: Dynamic multi-armed Bandits and Extreme value-based rewards for adaptive operator selection in evolutionary algorithms. In: Stützle, T. (ed.) LION 2009. LNCS, vol. 5851, pp. 176–190. Springer, Heidelberg (2009).  https://doi.org/10.1007/978-3-642-11169-3_13 CrossRefGoogle Scholar
  11. 11.
    Fialho, A., Da Costa, L., Schoenauer, M., Sebag, M.: Analyzing bandit-based adaptive operator selection mechanisms. AMAI 60, 25–64 (2010)MathSciNetzbMATHGoogle Scholar
  12. 12.
    García-Valdez, M., Trujillo, L., Merelo-Guérvos, J.J., Fernández-de-Vega, F.: Randomized parameter settings for heterogeneous workers in a pool-based evolutionary algorithm. In: Bartz-Beielstein, T., Branke, J., Filipič, B., Smith, J. (eds.) PPSN 2014. LNCS, vol. 8672, pp. 702–710. Springer, Cham (2014).  https://doi.org/10.1007/978-3-319-10762-2_69 Google Scholar
  13. 13.
    Goëffon, A., Lardeux, F., Saubion, F.: Simulating non-stationary operators in search algorithms. Appl. Soft Comput. 38, 257–268 (2016)CrossRefGoogle Scholar
  14. 14.
    Grefenstette, J.J.: Optimization of control parameters for genetic algorithms. IEEE Trans. Syst. Man Cybern. 16(1), 122–128 (1986)CrossRefGoogle Scholar
  15. 15.
    Jankee, C., Verel, S., Derbel, B., Fonlupt, C.: Distributed adaptive metaheuristic selection: comparisons of selection strategies. In: EA 2015, pp. 83–96 (2015)Google Scholar
  16. 16.
    Jankee, C., Verel, S., Derbel, B., Fonlupt, C.: A fitness cloud model for adaptive metaheuristic selection methods. In: Handl, J., Hart, E., Lewis, P.R., López-Ibáñez, M., Ochoa, G., Paechter, B. (eds.) PPSN 2016. LNCS, vol. 9921, pp. 80–90. Springer, Cham (2016).  https://doi.org/10.1007/978-3-319-45823-6_8 CrossRefGoogle Scholar
  17. 17.
    Kotthoff, L.: Algorithm selection for combinatorial search problems: a survey. AI Mag. 48–60 (2012)Google Scholar
  18. 18.
    Maturana, J., Fialho, Á., Saubion, F., Schoenauer, M., Sebag, M.: Extreme compass and dynamic multi-armed bandits for adaptive operator selection. In: CEC 2009, pp. 365–372. IEEE (2009)Google Scholar
  19. 19.
    Muniglia, M., Do, J.-M., Jean-Charles, L.P., Grard, H., Verel, S., David, S.: A multi-physics PWR model for the load following. In: ICAPP, April 2016Google Scholar
  20. 20.
    Sudholt, D.: Parallel evolutionary algorithms. In: Kacprzyk, J., Pedrycz, W. (eds.) Springer Handbook of Computational Intelligence, pp. 929–959. Springer, Heidelberg (2015).  https://doi.org/10.1007/978-3-662-43505-2_46 CrossRefGoogle Scholar
  21. 21.
    Tanabe, R., Fukunaga, A.: Evaluation of a randomized parameter setting strategy for island-model evolutionary algorithms. In: CEC, pp. 1263–1270 (2013)Google Scholar
  22. 22.
    Thierens, D.: An adaptive pursuit strategy for allocating operator probabilities. In: GECCO 2005, pp. 1539–1546 (2005)Google Scholar
  23. 23.
    Wessing, S., Rudolph, G., Menges, D.A.: Comparing asynchronous and synchronous parallelization of the SMS-EMOA. In: Handl, J., Hart, E., Lewis, P.R., López-Ibáñez, M., Ochoa, G., Paechter, B. (eds.) PPSN 2016. LNCS, vol. 9921, pp. 558–567. Springer, Cham (2016).  https://doi.org/10.1007/978-3-319-45823-6_52 CrossRefGoogle Scholar
  24. 24.
    Yagoubi, M., Schoenauer, M.: Asynchronous master/slave MOEAs and heterogeneous evaluation costs. In: GECCO, pp. 1007–1014 (2012)Google Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Christopher Jankee
    • 1
  • Sébastien Verel
    • 1
  • Bilel Derbel
    • 2
  • Cyril Fonlupt
    • 1
  1. 1.Université du Littoral Côte d’Opale, LISICCalaisFrance
  2. 2.Université Lille 1, LIFL, CNRS, INRIA LilleVilleneuve-d’AscqFrance

Personalised recommendations