Inter- and Intragenerational Mutation Shape Adaptation

Till now, only uncorrelated evolution strategies benefit from the strength of the directed mutation principle. It is thus a natural idea to enhance more powerful correlated evolution strategies with directed mutation, too. This work aims at porting this approach from the uncorrelated setting in classical evolution strategies to the correlated case as given in covariance matrix adaptation-evolution strategies. The main problem to be addressed here is the shape vector update. The shape vector controls the distribution’s skewness and can be updated intergenerationally as well as intragenerationally. Starting with an analogue to the intergenerational parameter update mechanics used in CMA-ES, we argue that an additional intragenerational update is of greater benefit. An appropriate heuristic will be presented and some experimental data of several test functions is provided.

[1]  G. Unter Rudolph Local Convergence Rates of Simple Evolutionary Algorithms with Cauchy Mutations , 1998 .

[2]  Xin Yao,et al.  Fast Evolution Strategies , 1997, Evolutionary Programming.

[3]  Stefan Berlik Directed Mutation by Means of the Skew-Normal Distribution , 2004, Fuzzy Days.

[4]  Thomas Bäck,et al.  A Survey of Evolution Strategies , 1991, ICGA.

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

[6]  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.

[7]  Anne Auger,et al.  LS-CMA-ES: A Second-Order Algorithm for Covariance Matrix Adaptation , 2004, PPSN.

[8]  Thomas Bäck,et al.  Evolutionary computation: comments on the history and current state , 1997, IEEE Trans. Evol. Comput..

[9]  Julian F. Miller,et al.  Genetic and Evolutionary Computation — GECCO 2003 , 2003, Lecture Notes in Computer Science.

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

[11]  Ingo Rechenberg,et al.  Evolutionsstrategie : Optimierung technischer Systeme nach Prinzipien der biologischen Evolution , 1973 .

[12]  Stefan Berlik A Step Size Preserving Directed Mutation Operator , 2004, GECCO.

[13]  Madjid Fathi,et al.  Multi-Objective Optimization Using Directed Mutation , 2005, IC-AI.

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