Foundations of Directed Mutation

Directed mutation abandons the so-called random mutation hypothesis postulating mutations to occur at random, regardless of fitness consequences to the resulting offspring. By introducing skewness into the mutation operators, bigger portions of offspring can be created in the area of higher fitness with respect to the elder and thus promising directions of the evolution path can be favored. The aim of this work is to present the foundations of directed mutation as well as different operators in one single place. Their characteristics will be presented and their advantages and disadvantages are discussed. Furthermore, an application scenario will be presented that shows how evolutionary algorithm and directed mutation can be applied in engineering design. In addition, some experimental results solving a real world optimization task in this scenario are provided. Finally some first, preliminary results of a multivariate skew distribution as mutation operator in a covariance matrix adaptation algorithm will be presented.

[1]  Günter Rudolph,et al.  Local convergence rates of simple evolutionary algorithms with Cauchy mutations , 1997, IEEE Trans. Evol. Comput..

[2]  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.

[3]  Kalyanmoy Deb,et al.  Multi-objective optimization using evolutionary algorithms , 2001, Wiley-Interscience series in systems and optimization.

[4]  M. Genton,et al.  A SKEW-SYMMETRIC REPRESENTATION OF MULTIVARIATE DISTRIBUTIONS , 2002 .

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

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

[7]  W. Vent,et al.  Rechenberg, Ingo, Evolutionsstrategie — Optimierung technischer Systeme nach Prinzipien der biologischen Evolution. 170 S. mit 36 Abb. Frommann‐Holzboog‐Verlag. Stuttgart 1973. Broschiert , 1975 .

[8]  A. Azzalini,et al.  The multivariate skew-normal distribution , 1996 .

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

[10]  R. Beaver,et al.  Skewed multivariate models related to hidden truncation and/or selective reporting , 2002 .

[11]  Hans-Paul Schwefel,et al.  Evolution and Optimum Seeking: The Sixth Generation , 1993 .

[12]  A. Azzalini.  A class of distributions which includes the normal ones , 1985 .

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

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

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

[16]  Thomas Bäck,et al.  An Overview of Evolutionary Algorithms for Parameter Optimization , 1993, Evolutionary Computation.