Analysis of Allele Distribution Dynamics in Different Genetic Algorithms
This chapter exemplarily points out how essential genetic information evolves during the runs of certain selected variants of a genetic algorithm. The discussed algorithmic enhancements to a standard genetic algorithm are motivated by Holland’s schema theory and the according building block hypothesis. The discussed offspring selection and the relevant alleles preserving genetic algorithm certify the survival of essential genetic information by supporting the survival of relevant alleles rather than the survival of above average chromosomes. This is achieved by defining the survival probability of a new child chromosome depending on the child’s fitness in comparison to the fitness values of its own parents. By this means the survival and expansion of essential building block information information is supported also for problem representations and algorithmic settings which do not fulfill the theoretical requirements of the schema theory. The properties of these GA variants are analyzed empirically. The selected analysis method assumes the knowledge of the unique globally optimal solution and is therefore restricted to rather theoretical considerations. The main aim of this chapter is to motivate and discuss the most important properties of the discussed algorithm variants in a rather intuitive way. Aspects for meaningful and practically more relevant generalizations as well as more sophisticated experimental analyses are indicated.
KeywordsGenetic Algorithm Crossover Operator Success Ratio Premature Convergence Offspring Selection
Unable to display preview. Download preview PDF.
- 2.Affenzeller, M., Wagner, S.: SASEGASA: A new generic parallel evolutionary algorithm for achieving highest quality results. Journal of Heuristics - Special Issue on New Advances on Parallel Meta-Heuristics for Complex Problems 10, 239–263 (2004)Google Scholar
- 3.Affenzeller, M., Wagner, S., Winkler, S.M.: Goal-oriented preservation of essential genetic information by offspring selection. In: Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2005), vol. 2, pp. 1595–1596. Association for Computing Machinery (ACM), New York (2005)CrossRefGoogle Scholar
- 6.Affenzeller, M., Beham, A., Kofler, M., Kronberger, G., Wagner, S., Winkler, S.M.: Metaheuristic Optimization. In: Hagenberg Research Software Engineering, pp. 103–155. Springer, Heidelberg (2009)Google Scholar
- 8.Beyer, H.G.: The Theory of Evolution Strategies. Springer, Heidelberg (2001)Google Scholar
- 9.Cavicchio, D.: Adaptive Search Using Simulated Evolution. PhD thesis, University of Michigan (1975)Google Scholar
- 10.DeJong, K.A.: An Analysis of the Behavior of a Class of Genetic Adaptive Systems. PhD thesis, University of Michigan (1975)Google Scholar
- 13.Holland, J.H.: Adaption in Natural and Artifical Systems. University of Michigan Press, Ann Arbor (1975)Google Scholar
- 16.Rechenberg, I.: Evolutionsstrategie. Friedrich Frommann Verlag (1973)Google Scholar
- 18.Schwefel, H.-P.: Numerische Optimierung von Computer-Modellen mittels der Evolutionsstrategie. Birkhäuser Verlag, Switzerland (1994)Google Scholar
- 19.Smith, R.E., Forrest, S., Perelson, A.S.: Population diversity in an immune systems model: Implications for genetic search. In: Foundations of Genetic Algorithms, vol. 2, pp. 153–166. Morgan Kaufmann Publishers, San Francisco (1993)Google Scholar
- 21.Wagner, S., Affenzeller, M.: SexualGA: Gender-specific selection for genetic algorithms. In: Callaos, N., Lesso, W., Hansen, E. (eds.) Proceedings of the 9th World Multi-Conference on Systemics, Cybernetics and Informatics (WMSCI 2005), vol. 4, pp. 76–81. International Institute of Informatics and Systemics (2005)Google Scholar