Comparing Asynchronous and Synchronous Parallelization of the SMS-EMOA

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9921)

Abstract

We experimentally compare synchronous and asynchronous parallelization of the SMS-EMOA. We find that asynchronous parallelization usually obtains a better speed-up and is more robust to fluctuations in the evaluation time of objective functions. Simultaneously, the solution quality of both methods only degrades slightly as against the sequential variant. We even consider it possible for the parallelization to improve the quality of the solution set on some multimodal problems.

Keywords

Asynchronous Synchronous Parallel Multiobjective Evolutionary Optimization 

References

  1. 1.
    Beume, N., Naujoks, B., Emmerich, M.: SMS-EMOA: multiobjective selection based on dominated hypervolume. Eur. J. Oper. Res. 181(3), 1653–1669 (2007)CrossRefMATHGoogle Scholar
  2. 2.
    Bringmann, K., Friedrich, T.: Don’t be greedy when calculating hypervolume contributions. In: Proceedings of the Tenth ACM SIGEVO Workshop on Foundations of Genetic Algorithms, FOGA 2009, pp. 103–112. ACM (2009)Google Scholar
  3. 3.
    Depolli, M., Trobec, R., Filipič, B.: Asynchronous master-slave parallelization of differential evolution for multi-objective optimization. Evol. Comput. 21(2), 261–291 (2012)CrossRefGoogle Scholar
  4. 4.
    Huband, S., Hingston, P., Barone, L., While, L.: A review of multiobjective test problems and a scalable test problem toolkit. IEEE Trans. Evol. Comput. 10(5), 477–506 (2006)CrossRefMATHGoogle Scholar
  5. 5.
    Klinkenberg, J.W., Emmerich, M.T.M., Deutz, A.H., Shir, O.M., Bäck, T.: A reduced-cost SMS-EMOA using kriging, self-adaptation, and parallelization. In: Ehrgott, M., Naujoks, B., Stewart, J.T., Wallenius, J. (eds.) Multiple Criteria Decision Making for Sustainable Energy and Transportation Systems. Lecture Notes in Economics and Mathematical Systems, vol. 634, pp. 301–311. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  6. 6.
    Märtens, M., Izzo, D.: The asynchronous island model and NSGA-II: study of a new migration operator and its performance. In: Proceedings of the 15th Annual Conference on Genetic and Evolutionary Computation, GECCO 2013, pp. 1173–1180. ACM (2013)Google Scholar
  7. 7.
    Menges, D.A., Wessing, S., Rudolph, G.: Asynchrone Parallelisierung des SMS-EMOA zur Parameteroptimierung von mobilen Robotern. In: Hoffmann, F., Hüllermeier, E. (eds.) Proceedings 25, Workshop Computational Intelligence. Schriftenreihe des Instituts für Angewandte Informatik/Automatisierungstechnik, vol. 54, pp. 47–65. KIT Scientific Publishing (2015). (in German)Google Scholar
  8. 8.
    Schütze, O., Esquivel, X., Lara, A., Coello Coello, C.A.: Using the averaged Hausdorff distance as a performance measure in evolutionary multiobjective optimization. IEEE Trans. Evol. Comput. 16(4), 504–522 (2012)CrossRefGoogle Scholar
  9. 9.
    Wessing, S.: evoalgos: modular evolutionary algorithms (2016). Python package version 0.3. https://pypi.python.org/pypi/evoalgos
  10. 10.
    Wessing, S.: optproblems: infrastructure to define optimization problems and some test problems for black-box optimization (2016). Python package version 0.8. https://pypi.python.org/pypi/optproblems

Copyright information

© Springer International Publishing AG 2016

Authors and Affiliations

  • Simon Wessing
    • 1
  • Günter Rudolph
    • 1
  • Dino A. Menges
    • 2
  1. 1.Computer Science DepartmentTechnische Universität DortmundDortmundGermany
  2. 2.Adept Technology GmbHDortmundGermany

Personalised recommendations