Maximum Likelihood-Based Online Adaptation of Hyper-Parameters in CMA-ES

  • Ilya Loshchilov
  • Marc Schoenauer
  • Michèle Sebag
  • Nikolaus Hansen
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8672)


The Covariance Matrix Adaptation Evolution Strategy (CMA-ES) is widely accepted as a robust derivative-free continuous optimization algorithm for non-linear and non-convex optimization problems. CMA-ES is well known to be almost parameterless, meaning that only one hyper-parameter, the population size, is proposed to be tuned by the user. In this paper, we propose a principled approach called self-CMA-ES to achieve the online adaptation of CMA-ES hyper-parameters in order to improve its overall performance. Experimental results show that for larger-than-default population size, the default settings of hyper-parameters of CMA-ES are far from being optimal, and that self-CMA-ES allows for dynamically approaching optimal settings.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Auger, A., Hansen, N.: A Restart CMA Evolution Strategy With Increasing Population Size. In: IEEE Congress on Evolutionary Computation, pp. 1769–1776. IEEE Press (2005)Google Scholar
  2. 2.
    Beyer, H.-G., Hellwig, M.: Controlling population size and mutation strength by meta-es under fitness noise. In: Proceedings of the Twelfth Workshop on Foundations of Genetic Algorithms XII, FOGA XII 2013, pp. 11–24. ACM (2013)Google Scholar
  3. 3.
    Hansen, N.: Benchmarking a BI-population CMA-ES on the BBOB-2009 function testbed. In: GECCO Companion, pp. 2389–2396 (2009)Google Scholar
  4. 4.
    Hansen, N., Auger, A., Finck, S., Ros, R.: Real-Parameter Black-Box Optimization Benchmarking 2010: Experimental Setup. Technical report, INRIA (2010)Google Scholar
  5. 5.
    Hansen, N., Müller, S., Koumoutsakos, P.: Reducing the time complexity of the derandomized evolution strategy with covariance matrix adaptation (CMA-ES). Evolutionary Computation 11(1), 1–18 (2003)CrossRefGoogle Scholar
  6. 6.
    Hansen, N., Ostermeier, A.: Adapting Arbitrary Normal Mutation Distributions in Evolution Strategies: The Covariance Matrix Adaptation. In: International Conference on Evolutionary Computation, pp. 312–317 (1996)Google Scholar
  7. 7.
    Hansen, N., Ostermeier, A.: Completely Derandomized Self-Adaptation in Evolution Strategies. Evolutionary Computation 9(2), 159–195 (2001)CrossRefGoogle Scholar
  8. 8.
    Hansen, N., Ros, R.: Benchmarking a weighted negative covariance matrix update on the BBOB-2010 noisy testbed. In: GECCO 2010: Proceedings of the 12th Annual Conference Comp on Genetic and Evolutionary Computation, pp. 1681–1688. ACM, New York (2010)Google Scholar
  9. 9.
    Hoffmann, F., Holemann, S.: Controlled Model Assisted Evolution Strategy with Adaptive Preselection. In: International Symposium on Evolving Fuzzy Systems, pp. 182–187. IEEE (2006)Google Scholar
  10. 10.
    Igel, C., Hüsken, M.: Empirical evaluation of the improved rprop learning algorithms. Neurocomputing 50, 105–123 (2003)CrossRefMATHGoogle Scholar
  11. 11.
    Liao, T., Stützle, T.: Benchmark results for a simple hybrid algorithm on the CEC 2013 benchmark set for real-parameter optimization. In: IEEE Congress on Evolutionary Computation (CEC), pp. 1938–1944. IEEE Press (2013)Google Scholar
  12. 12.
    Loshchilov, I., Schoenauer, M., Sebag, M.: Alternative Restart Strategies for CMA-ES. In: Coello, C.A.C., Cutello, V., Deb, K., Forrest, S., Nicosia, G., Pavone, M. (eds.) PPSN 2012, Part I. LNCS, vol. 7491, pp. 296–305. Springer, Heidelberg (2012)CrossRefGoogle Scholar
  13. 13.
    Loshchilov, I., Schoenauer, M., Sebag, M.: Self-Adaptive Surrogate-Assisted Covariance Matrix Adaptation Evolution Strategy. In: Genetic and Evolutionary Computation Conference (GECCO), pp. 321–328. ACM Press (July 2012)Google Scholar
  14. 14.
    Loshchilov, I., Schoenauer, M., Sebag, M.: Intensive Surrogate Model Exploitation in Self-adaptive Surrogate-assisted CMA-ES (saACM-ES). In: Genetic and Evolutionary Computation Conference, pp. 439–446. ACM (2013)Google Scholar
  15. 15.
    Schaul, T.: Comparing natural evolution strategies to bipop-cma-es on noiseless and noisy black-box optimization testbeds. In: Genetic and Evolutionary Computation Conference Companion, pp. 237–244. ACM (2012)Google Scholar
  16. 16.
    Smit, S., Eiben, A.: Beating the ‘world champion’ Evolutionary Algorithm via REVAC Tuning. IEEE Congress on Evolutionary Computation, 1–8 (2010)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Ilya Loshchilov
    • 1
  • Marc Schoenauer
    • 2
    • 3
  • Michèle Sebag
    • 3
    • 2
  • Nikolaus Hansen
    • 2
    • 3
  1. 1.Laboratory of Intelligent SystemsÉcole Polytechnique Fédérale de LausanneSwitzerland
  2. 2.TAO Project-teamINRIA SaclayÎle-de-FranceFrance
  3. 3.Laboratoire de Recherche en Informatique (UMR CNRS 8623)Université Paris-SudOrsay CedexFrance

Personalised recommendations