Advertisement

Reproducible Roulette Wheel Sampling for Message Passing Environments

  • Balazs Nemeth
  • Tom Haber
  • Jori Liesenborgs
  • Wim Lamotte
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10861)

Abstract

Roulette Wheel Sampling, sometimes referred to as Fitness Proportionate Selection, is a method to sample from a set of objects each with an associated weight. This paper introduces a distributed version of the method designed for message passing environments. Theoretical bounds are derived to show that the presented method has better scalability than naive approaches. This is verified empirically on a test cluster, where improved speedup is measured. In all tested configurations, the presented method performs better than naive approaches. Through a renumbering step, communication volume is minimized. This step also ensures reproducibility regardless of the underlying architecture.

Keywords

Genetic algorithms Roulette wheel selection Sequential Monte Carlo HPC Message passing 

Notes

Acknowledgments

Part of the work presented in this paper was funded by Johnson & Johnson.

References

  1. 1.
    de Freitas, N., Gordon, N., Doucet, A. (eds.): Sequential Monte Carlo Methods in Practice. Springer, Heidelberg (2001).  https://doi.org/10.1007/978-1-4757-3437-9CrossRefzbMATHGoogle Scholar
  2. 2.
    Blelloch, G.E.: Prefix sums and their applications. Technical report. Synthesis of Parallel Algorithms (1990)Google Scholar
  3. 3.
    Cant-Paz, E.: A survey of parallel genetic algorithms. Calculateurs Paralleles et Reseaux Syst. Repartis 10(2), 141–171 (1998)Google Scholar
  4. 4.
    Goldberg, D.E.: Genetic Algorithms. Pearson Education India, Noida (2006)Google Scholar
  5. 5.
    Kumar, V.: Introduction to Parallel Computing, 2nd edn. Addison-Wesley Longman Publishing Co., Inc., Boston (2002)Google Scholar
  6. 6.
    Lipowski, A., Lipowska, D.: Roulette-wheel selection via stochastic acceptance. Phys. A Stat. Mech. Appl. 391(6), 2193–2196 (2012)CrossRefGoogle Scholar
  7. 7.
    Moral, P.D., Jasra, A., Law, K.J.H., Zhou, Y.: Multilevel Sequential Monte Carlo samplers for normalizing constants. ACM Trans. Model. Comput. Simul. 27(3), 20:1–20:22 (2017)MathSciNetCrossRefGoogle Scholar
  8. 8.
    Salmon, J.K., Moraes, M.A., Dror, R.O., Shaw, D.E.: Parallel random numbers: as easy as 1, 2, 3. In: Proceedings of 2011 International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2011, pp. 16:1–16:12. ACM, New York (2011)Google Scholar
  9. 9.
    Li, S., Hu, J., Cheng, X., Zhao, C.: Asynchronous work stealing on distributed memory systems, pp. 198–202. IEEE, February 2013Google Scholar
  10. 10.
    Vose, M.D.: A linear algorithm for generating random numbers with a given distribution. IEEE Trans. Softw. Eng. 17(9), 972–975 (1991)MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Balazs Nemeth
    • 1
  • Tom Haber
    • 1
    • 2
  • Jori Liesenborgs
    • 1
  • Wim Lamotte
    • 1
  1. 1.Expertise Centre for Digital MediaDiepenbeekBelgium
  2. 2.Exascience LabImecLeuvenBelgium

Personalised recommendations