Siddhartha Shakya
ABSTRACT
This paper presents an empirical cost-benefit analysis of an algorithm called Distribution Estimation Using MRF with direct sampling (DEUMd). DEUMd belongs to the family of Estimation of Distribution Algorithm (EDA). Particularly it is a univariate EDA. DEUMd uses a computationally more expensive model to estimate the probability distribution than other univariate EDAs. We investigate the performance of DEUMd in a range of optimization problem. Our experiments shows a better performance (in terms of the number of fitness evaluation needed by the algorithm to find a solution and the quality of the solution) of DEUMd on most of the problems analysed in this paper in comparison to that of other univariate EDAs. We conclude that use of a Markov Network in a univariate EDA can be of net benefit in defined set of circumstances.
- S. Baluja. Population-based incremental learning: A method for integrating genetic search based function optimization and competitive learning,. Technical Report CMU-CS-94-163, Pittsburgh, PA, 1994. Google Scholar
Digital Library
- S. Baluja and S. Davies. Using optimal dependency-trees for combinatorial optimization: Learning the structure of the search space. In Proceedings of the 1997 International Conference on Machine Learning 1997. Google ScholarDigital Library
- D. F. Brown, A. B. Garmendia-Doval, and J. A. W. McCall. MarkovRandom Field Modelling of Royal Road Genetic Algorithms. Lecture Notes in Computer Science 2310: 65--78, January 2002. Google ScholarDigital Library
- L. Davis, editor. Handbook of Genetic Algorithms Van Nostrand Reinhold, New York, 1991.Google Scholar
- L. delaOssa, J. A. Gámez, and J. M. Puerta. Migration of Probability Models Instead of Individuals: An Alternative When Applying the Island Model to EDAs. In Parallel Problem Solving from Nature VIII pages 242--252. Springer, 2004.Google Scholar
- D. Goldberg. Genetic Algorithms in Search, Optimization, and Machine Learning Addison-Wesley, 1989. Google ScholarDigital Library
- G. Golub and C. Van Loan. Matrix Computations Baltimore, MD, second edition, 1989.Google Scholar
- G. R. Harik, F. G. Lobo, and D. E. Goldberg. The compact genetic algorithm. IEEE-EC 3(4): 287, November 1999. Google ScholarDigital Library
- H. H. Hoos and T. Stutzle. Towards a characterisation of the behaviour of stochastic local search algorithms for SAT. Artificial Intelligence 112(1-2): 213--232, 1999. Google ScholarDigital Library
- P. Larrañaga, R. Etxeberria, J. Lozano, and J. Peña. Optimization by learning and simulation of bayesian and gaussian networks. Technical Report EHU-KZAA-IK-4/99, University of the Basque Country, 1999.Google Scholar
- P. Larrañaga and J. A. Lozano. Estimation of Distribution Algorithms: A New Tool for Evolutionary Computation Kluwer Academic Publishers, 2001. Google ScholarDigital Library
- S. Z. Li. Markov Random Field modeling in computer vision Springer-Verlag, 1995. Google ScholarDigital Library
- H. Mühlenbein. Evolutionary algorithms and the boltzmann distribution. In Foundations of Genetic Algorithms (FOGA2002).Google Scholar
- H. Mühlenbein. The science of breeding and its application to the breeder genetic algorithm. Evolutionary Computation 1: pp. 335--360., 1994.Google ScholarDigital Library
- H. Mühlenbein and R. Höns. A theoretical and experimental investigation of estimation of distribution algorithms. In Optimization by Building and Using Probabilistic Models (OBUPM-2004).Google Scholar
- H. Mühlenbein and T. Mahnig. FDA -A scalable evolutionary algorithm for the optimization of additively decomposed functions. Evolutionary Computation 7(4): 353--376, 1999. Google ScholarDigital Library
- H. Mühlenbein and G. Paas. From recombination of genes to the estimation of distributions: I. binary parameters. In H. -M. Voigt, W. Ebeling, I. Rechenberg, and H. -P. Schwefel, editors, Parallel Problem Solving from Nature - PPSN IV pages 178--187, Berlin, 1996. Springer. Google ScholarDigital Library
- I. Murray and Z. Ghahramani. Bayesian Learning in Undirected Graphical Models: Approximate MCMC algorithms. In Twentieth Conference on Uncertainty in Artificial Intelligence (UAI 2004), Ban--, Canada, 8-11 July 2004. Google ScholarDigital Library
- M. Pelikan. Bayesian optimization algorithm: From single level to hierarchy PhD thesis, University of Illinois at Urbana-Champaign, Urbana, IL, 2002. Also IlliGAL Report No. 2002023. Google ScholarDigital Library
- M. Pelikan and D. E. Goldberg. Hierarchical BOA solves Ising spin glasses and MAXSAT. Proceedings of the Genetic and Evolutionary Computation Conference (GECCO-2003), pages 1271--1282, 2003. Also IlliGAL Report No. 2003001. Google ScholarDigital Library
- M. Pelikan, D. E. Goldberg, and F. Lobo. A survey of optimization by building and using probabilistic models. Technical Report 99018, Illinois Genetic Algorithms Lab, UIUC, Urbana, IL, 1999.Google Scholar
- W. H. Press, S. A. Teukolsky, W. T. Vetterling, and B. P. Flannery. Numerical Recipes in C: The Art of Scientific Computing Cambridge University Press, Cambridge, UK, 2nd edition, 1993. Google ScholarDigital Library
- S. Shakya, J. McCall, and D. Brown. Estimating the distribution in an EDA. In B. Ribeiro, R. F. Albrechet, A. Dobnikar, D. W. Pearson, and N. C. Steele, editors, In proceedings of the International Conference on Adaptive and Natural computiNG Algorithms (ICANNGA 2005), pages 202--205, Coimbra, Portugal, 2005. Springer-Verlag, Wien.Google Scholar
- S. K. Shakya, J. A. W. McCall, and D. F. Brown. Updating the probability vector using MRF technique for a Univariate EDA. In E. Onaindia and S. Staab, editors, Proceedings of the Second Starting AI Researchers' Symposium pages 15--25, Valencia, Spain, 2004. IOS press.Google Scholar
- R. Suda and S. Kuriyama. Another preprocessing algorithm for generalized one-dimensional fast multipole method. Journal of Computational Physics 195: 790--803, 2004. Google ScholarDigital Library
Index Terms
- Using a Markov network model in a univariate EDA: an empirical cost-benefit analysis
Recommendations
Solving the MAXSAT problem using a multivariate EDA based on Markov networks
GECCO '07: Proceedings of the 9th annual conference companion on Genetic and evolutionary computationMarkov Networks (also known as Markov Random Fields) have been proposed as a new approach to probabilistic modelling in Estimation of Distribution Algorithms (EDAs). An EDA employing this approach called Distribution Estimation Using Markov Networks (...
Evolutionary algorithms and elliptical copulas applied to continuous optimization problems
Estimation of Distribution Algorithms (EDAs) constitutes a class of evolutionary algorithms that can extract and exploit knowledge acquired throughout the optimization process. The most critical step in the EDAs is the estimation of the joint ...
Probabilistic model-building genetic algorithms
GECCO '07: Proceedings of the 9th annual conference companion on Genetic and evolutionary computationProbabilistic model-building algorithms (PMBGAs) replace traditional variation of genetic and evolutionary algorithms by (1) building a probabilistic model of promising solutions and (2) sampling the built model to generate new candidate solutions. ...
Comments