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

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.

[1]  Petros Koumoutsakos,et al.  Reducing the Time Complexity of the Derandomized Evolution Strategy with Covariance Matrix Adaptation (CMA-ES) , 2003, Evolutionary Computation.

[2]  A. E. Eiben,et al.  Beating the ‘world champion’ evolutionary algorithm via REVAC tuning , 2010, IEEE Congress on Evolutionary Computation.

[3]  Michèle Sebag,et al.  Self-adaptive surrogate-assisted covariance matrix adaptation evolution strategy , 2012, GECCO '12.

[4]  Christian Igel,et al.  Empirical evaluation of the improved Rprop learning algorithms , 2003, Neurocomputing.

[5]  Nikolaus Hansen,et al.  Adapting arbitrary normal mutation distributions in evolution strategies: the covariance matrix adaptation , 1996, Proceedings of IEEE International Conference on Evolutionary Computation.

[6]  Hans-Georg Beyer,et al.  Controlling population size and mutation strength by Meta-ES under fitness noise , 2013, FOGA XII '13.

[7]  Nikolaus Hansen,et al.  A restart CMA evolution strategy with increasing population size , 2005, 2005 IEEE Congress on Evolutionary Computation.

[8]  Tom Schaul,et al.  Comparing natural evolution strategies to BIPOP-CMA-ES on noiseless and noisy black-box optimization testbeds , 2012, GECCO '12.

[9]  Thomas Stützle,et al.  Benchmark results for a simple hybrid algorithm on the CEC 2013 benchmark set for real-parameter optimization , 2013, 2013 IEEE Congress on Evolutionary Computation.

[10]  Nikolaus Hansen,et al.  Benchmarking a BI-population CMA-ES on the BBOB-2009 function testbed , 2009, GECCO '09.

[11]  Michèle Sebag,et al.  Intensive surrogate model exploitation in self-adaptive surrogate-assisted cma-es (saacm-es) , 2013, GECCO '13.

[12]  F. Hoffmann,et al.  Controlled Model Assisted Evolution Strategy with Adaptive Preselection , 2006, 2006 International Symposium on Evolving Fuzzy Systems.

[13]  Raymond Ros,et al.  Benchmarking a weighted negative covariance matrix update on the BBOB-2010 noiseless testbed , 2010, GECCO '10.

[14]  Raymond Ros,et al.  Real-Parameter Black-Box Optimization Benchmarking 2009: Experimental Setup , 2009 .

[15]  Nikolaus Hansen,et al.  Completely Derandomized Self-Adaptation in Evolution Strategies , 2001, Evolutionary Computation.

[16]  Michèle Sebag,et al.  Alternative Restart Strategies for CMA-ES , 2012, PPSN.