Online vs. Offline ANOVA Use on Evolutionary Algorithms

One of the main drawbacks of evolutionary algorithms is their great amount of parameters. Every step to lower this quantity is a step in the right direction. Automatic control of variation operators application rates during the run of an evolutionary algorithm is a desirable feature for two reasons: we are lowering the number of parameters of the algorithm and making it able to react changes in the conditions of the problem. In this paper, a dynamic breeder able to adapt the operators application rates over time following the evolutionary process is proposed. The decision to raise or to lower every rate is based on ANOVA to be sure of statistical significant.

[1]  Terry Jones,et al.  Crossover, Macromutationand, and Population-Based Search , 1995, ICGA.

[2]  Jim Smith,et al.  Operator and parameter adaptation in genetic algorithms , 1997, Soft Comput..

[3]  Christian Igel,et al.  Balancing Learning And Evolution , 2002, GECCO.

[4]  Héctor Pomares,et al.  Statistical analysis of the main parameters involved in the design of a genetic algorithm , 2002, IEEE Trans. Syst. Man Cybern. Part C.

[5]  Seppo J. Ovaska,et al.  A general framework for statistical performance comparison of evolutionary computation algorithms , 2006, Inf. Sci..

[6]  Ignacio Rojas,et al.  Statistical analysis of the parameters of a neuro-genetic algorithm , 2002, IEEE Trans. Neural Networks.

[7]  Marc Toussaint,et al.  Self-adaptive exploration in evolutionary search , 2001, ArXiv.

[8]  Christian Igel,et al.  Operator adaptation in structure optimization of neural networks , 2001 .

[9]  R. Mead,et al.  The Design of Experiments: Statistical Principles for Practical Applications. , 1989 .

[10]  Zbigniew Michalewicz,et al.  Parameter Control in Evolutionary Algorithms , 2007, Parameter Setting in Evolutionary Algorithms.

[11]  Zbigniew Michalewicz,et al.  Parameter control in evolutionary algorithms , 1999, IEEE Trans. Evol. Comput..

[12]  Kenneth Alan De Jong,et al.  An analysis of the behavior of a class of genetic adaptive systems. , 1975 .

[13]  G. Casella,et al.  Statistical Inference , 2003, Encyclopedia of Social Network Analysis and Mining.

[14]  John J. Grefenstette,et al.  Optimization of Control Parameters for Genetic Algorithms , 1986, IEEE Transactions on Systems, Man, and Cybernetics.

[15]  Zbigniew Michalewicz,et al.  Genetic Algorithms + Data Structures = Evolution Programs , 1996, Springer Berlin Heidelberg.

[16]  Marimuthu Palaniswami,et al.  Computational Intelligence: A Dynamic System Perspective , 1995 .