Si-Cheng Chen
ABSTRACT
This paper investigates the difficulty of linkage learning, an essential core, in EDAs. Specifically, it examines allelicpairwise independent functions including the parity, paritywith-trap, and Walsh-code functions. While the parity function was believed to be difficult for EDAs in previous work, our experiments indicate that it can be solved by CGA within a polynomial number of function evaluations to the problem size. Consequently, the apparently difficult paritywith-trap function can be easily solved by ECGA, even though the linkage model is incorrect. A convergence model for CGA on the parity function is also derived to verify and support the empirical findings. Finally, this paper proposes a socalled Walsh-code function, which is more difficult than the parity function. Although the proposed function does deceive the linkage-learning mechanism in most EDAs, EDAs are still able to solve it to some extent.
- D. H. Ackley. A connectionist machine for genetic hillclimbing. Kluwer Academic Publishers, Norwell, MA, USA, 1987. Google Scholar
Digital Library
- S. Baluja. Population-based incremental learning: A method for integrating genetic search based function optimization and competitive learning. Technical report, Carnegie Mellon University, Pittsburgh, PA, USA, 1994. Google ScholarDigital Library
- J. S. D. Bonet, C. L. Isbell, and P. Viola. Mimic: Finding optima by estimating probability densities. In M. J. M.C. Mozer and T. Petsche, editors, Advances in Neural Information Processing Systems, volume 9, page 424. The MIT Press, 1997.Google Scholar
- D. Coffin and R. Smith. The limitations of distribution sampling for linkage learning. In Evolutionary Computation, 2007. CEC 2007. IEEE Congress on, pages 364--369, Sept. 2007.Google ScholarCross Ref
- C. Echegoyen, R. Santana, J. A. Lozano, and P. Larranaga. The impact of exact probabilistic learning algorithms in edas based on bayesian networks. Linkage in Evolutionary Computation, pages 109--139, 2008.Google ScholarCross Ref
- L. Emmendorfer and A. Pozo. A clustering-based approach for linkage learning applied to multimodal optimization. Linkage in Evolutionary Computation, pages 225--248, 2008.Google ScholarCross Ref
- R. Etxeberria and P. Larranaga. Global optimization using bayesian networks. In Second Symposium on Artificial Intelligence (CIMAF-99), pages 332--339, 1999.Google Scholar
- S. Forrest and M. Mitchell. What makes a problem hard for a genetic algorithm? some anomalous results and their explanation. In Machine Learning, pages 285--319, 1993. Google ScholarDigital Library
- D. E. Goldberg. Genetic algorithms and walsh functions: Part i, a gentle introduction. In Complex Systems, volume 3, pages 129 -- 152, 1989.Google Scholar
- D. E. Goldberg. The Design of Innovation: Lessons from and for Competent Genetic Algorithms, chapter 12, pages 187--216. Kluwer Academic Publishers, Norwell, MA, USA, 2002. Google ScholarDigital Library
- D. E. Goldberg, K. Deb, and J. H. Clark. Genetic algorithms, noise, and the sizing of populations. In Complex Systems, volume 6, pages 333 -- 362, 1992.Google Scholar
- G. Harik. Linkage learning via probabilistic modeling in the ecga. Technical report, IlliGAL, University of Illinois at Urbana-Champaign, 1999.Google Scholar
- G. Harik, E. Cantu-Paz, D. E. Goldberg, and B. L. Miller. The gambler's ruin problem, genetic algorithms, and the sizing of populations. Evolutionary Computation, 7(3):231--253, 1999. Google ScholarDigital Library
- G. Harik, F. Lobo, and D. Goldberg. The compact genetic algorithm. In Evolutionary Computation, 2007. CEC 2007. IEEE Congress on, pages 523--528, May 1998.Google ScholarCross Ref
- P. Larranaga and J. A. Lozano, editors. Estimation of Distribution Algorithms. A New Tool for Evolutionary Computation. Kluwer Academic Publishers, 2002. Google ScholarDigital Library
- C. Lima, M. Pelikan, D. Goldberg, F. Lobo, K. Sastry, and M. Hauschild. Influence of selection and replacement strategies on linkage learning in boa. In Evolutionary Computation, 2007. CEC 2007. IEEE Congress on, pages 1083--1090, Sept. 2007Google ScholarCross Ref
- H. M¨uhlenbein and G. Paass. From recombination of genes to the estimation of distributions i. binary parameters. In PPSN IV: Proceedings of the 4th International Conference on Parallel Problem Solving from Nature, pages 178--187, London, UK, 1996. Springer-Verlag. Google ScholarDigital Library
- M. Pelikan and D. E. Goldberg. Escaping hierarchical traps with competent genetic algorithms. In Proceedings of the Genetic and Evolutionary Computation Conference (GECCO2001), pages 511--518. Morgan Kaufmann, 2001.Google Scholar
- M. Pelikan, D. E. Goldberg, and E. Cantu-Paz. Boa: The bayesian optimization algorithm. In Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-99), pages 525--532, 1999.Google Scholar
- M. Pelikan, D. E. Goldberg, and F. G. Lobo. A survey of optimization by building and using probabilistic models. Comput. Optim. Appl., 21(1):5--20, 2002. Google ScholarDigital Library
- M. Pelikan and H. Muhlenbein. The bivariate marginal distribution algorithm. In R. Roy, T. Furuhashi, and P. Chawdhry, editors, Advances in Soft Computing - Engineering Design and Manufacturing, pages 521--535, London, 1999.Google Scholar
- M. Tsuji, M. Munetomo, and K. Akama. Modeling dependencies of loci with string classification according to fitness differences. In Genetic and Evolutionary Computation (GECCO 2004), pages 246--257, 2004.Google ScholarCross Ref
- H. Wu and J. L. Shapiro. Does overfitting affect performance in estimation of distribution algorithms. In GECCO '06: Proceedings of the 8th annual conference on Genetic and evolutionary computation, pages 433--434, New York, NY, USA, 2006. ACM. Google ScholarDigital Library
- T.-L. Yu, D. E. Goldberg, A. Yassine, and Y.-P. Chen. A genetic algorithm design inspired by organization theory: a pilot study of a dependency structure matrix driven genetic algorithm. Technical report, IlliGAL, University of Illinois at Urbana-Champaign, Urbana, IL, 2003.Google Scholar
Index Terms
- Difficulty of linkage learning in estimation of distribution algorithms
Recommendations
Sub-structural niching in estimation of distribution algorithms
GECCO '05: Proceedings of the 7th annual conference on Genetic and evolutionary computationWe propose a sub-structural niching method that fully exploits the problem decomposition capability of linkage-learning methods such as the estimation distribution algorithms and concentrate on maintaining diversity at the sub-structural level. The ...
A linkage-learning niching in estimation of distribution algorithm
GECCO '12: Proceedings of the 14th annual conference companion on Genetic and evolutionary computationThis work proposes a linkage-learning niching method that improves the capability of estimation of distribution algorithms (EDAs) on reducing spurious linkages which increase problems difficulty. Concatenated parity function (CPF), a class of allelic ...
A surrogate-assisted linkage inference approach in genetic algorithms
GECCO '11: Proceedings of the 13th annual conference on Genetic and evolutionary computationLinkage in terms of genetic algorithms is a measure of interdependence of groups of genes. When linkage exists, the fitness contribution of one gene depends on the allele setting of another. Our approach, a surrogate-assisted linkage inference genetic ...
Comments