Abstract
A maximum entropy approach is used to derive a set of equations describing the evolution of a genetic algorithm involving crossover, mutation and selection. The problem is formulated in terms of cumulants of the fitness distribution. Applying this method to very simple problems, the dynamics of the genetic algorithm can be reduced to a set of nonlinear coupled difference equations which give good results when truncated to four variables. These equations correctly predict the best fitness after convergence as a function of the genetic search parameters. Suggestions concerning annealing rates for time-dependent search parameters are also discussed.
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
This is a preview of subscription content, log in via an institution.
Preview
Unable to display preview. Download preview PDF.
References
John H. Holland. Adaptation in Natural and Artificial Systems. University of Michigan Press (Ann Arbor), 1975.
David E. Goldberg. Genetic algorithms and walsh functions: Part II, deception and its analysis. Complex Systems, 3:153–171, 1990.
A. Nix and Michael D. Vose. Modeling genetic algorithms with markov chains. Annals of Mathematics and Artificial Intelligence, 5:79–88, 1991.
H. Mühlenbein. Evolution in space and time — the parallel genetic algorithm. In G. Rawlins., editor, Foundations of Genetic Algorithms. Morgan-Kaufman, 1991.
Darrell Whitley. An executable model of a simple genetic algorithm. In L. Darrel Whitley, editor, Foundations of Genetic Algorithms 2. Morgan Kaufmann (San Mateo), 1993.
John J. Grefenstette. Deception considered harmful. In L. Darrel Whitley, editor, Foundations of Genetic Algorithms 2. Morgan Kaufmann (San Mateo), 1993.
A. Prügel-Bennett and J. L. Shapiro. An analysis of genetic algorithms using statistical mechanics. Phys. Rev. Letts., 72(9):1305–1309, 1994.
J. L. Shapiro, A. Prügel-Bennett, and M. Rattray. A statistical mechanical formulation of the dynamics of genetic algorithms. In Terence C. Fogarty, editor, Evolutionary Computing: AISB Workshop, Leeds, U.K., April 1994, Selected Papers, volume 864 of Lecture Notes in Computer Science, pages 17–27. Springer-Verlag, 1994.
A. Prügel-Bennett and J. L. Shapiro. Dynamics of genetic algorithms for the ising spin-glass chain.
Thomas Cover and Joy Thomas. Elements of Information Theory. Wiley, 1991.
Magnus Rattray. An analysis of a genetic algorithm training the binary perceptron. Master's thesis, University of Manchester, 1995.
Magnus Rattray. An analysis of a genetic algorithm solving the subset-sum problem. In preparation, 1995.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1995 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Shapiro, J.L., Prügel-Bennett, A. (1995). Maximum entropy analysis of genetic algorithm operators. In: Fogarty, T.C. (eds) Evolutionary Computing. AISB EC 1995. Lecture Notes in Computer Science, vol 993. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-60469-3_21
Download citation
DOI: https://doi.org/10.1007/3-540-60469-3_21
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-60469-3
Online ISBN: 978-3-540-47515-6
eBook Packages: Springer Book Archive