The CMA Evolution Strategy: A Comparing Review

Derived from the concept of self-adaptation in evolution strategies, the CMA (Covariance Matrix Adaptation) adapts the covariance matrix of a multi-variate normal search distribution. The CMA was originally designed to perform well with small populations. In this review, the argument starts out with large population sizes, reflecting recent extensions of the CMA algorithm. Commonalities and differences to continuous Estimation of Distribution Algorithms are analyzed. The aspects of reliability of the estimation, overall step size control, and independence from the coordinate system (invariance) become particularly important in small populations sizes. Consequently, performing the adaptation task with small populations is more intricate.

[1]  Ingo Rechenberg,et al.  Evolutionsstrategie '94 , 1994, Werkstatt Bionik und Evolutionstechnik.

[2]  Hans-Paul Schwefel,et al.  Evolution and optimum seeking , 1995, Sixth-generation computer technology series.

[3]  Rich Caruana,et al.  Removing the Genetics from the Standard Genetic Algorithm , 1995, ICML.

[4]  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.

[5]  N. Hansen,et al.  Convergence Properties of Evolution Strategies with the Derandomized Covariance Matrix Adaptation: T , 1997 .

[6]  Michèle Sebag,et al.  Extending Population-Based Incremental Learning to Continuous Search Spaces , 1998, PPSN.

[7]  Nikolaus Hansen,et al.  Verallgemeinerte individuelle Schrittweitenregelung in der Evolutionsstrategie , 1998 .

[8]  Dirk Thierens,et al.  Expanding from Discrete to Continuous Estimation of Distribution Algorithms: The IDEA , 2000, PPSN.

[9]  Nikolaus Hansen,et al.  Invariance, Self-Adaptation and Correlated Mutations and Evolution Strategies , 2000, PPSN.

[10]  J. A. Lozano,et al.  Estimation of Distribution Algorithms: A New Tool for Evolutionary Computation , 2001 .

[11]  Kalyanmoy Deb,et al.  On self-adaptive features in real-parameter evolutionary algorithms , 2001, IEEE Trans. Evol. Comput..

[12]  Hans-Georg Beyer,et al.  The Theory of Evolution Strategies , 2001, Natural Computing Series.

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

[14]  Pedro Larrañaga,et al.  A Review on Estimation of Distribution Algorithms , 2002, Estimation of Distribution Algorithms.

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

[16]  Hans-Georg Beyer,et al.  Qualms Regarding the Optimality of Cumulative Path Length Control in CSA/CMA-Evolution Strategies , 2003, Evolutionary Computation.

[17]  Marcus Gallagher,et al.  Playing in continuous spaces: some analysis and extension of population-based incremental learning , 2003, The 2003 Congress on Evolutionary Computation, 2003. CEC '03..

[18]  Petros Koumoutsakos,et al.  Learning probability distributions in continuous evolutionary algorithms – a comparative review , 2004, Natural Computing.

[19]  Nikolaus Hansen,et al.  Evaluating the CMA Evolution Strategy on Multimodal Test Functions , 2004, PPSN.