Abstract
This paper presents a Markov random field (MRF) approach to estimating and sampling the probability distribution in populations of solutions. The approach is used to define a class of algorithms under the general heading distribution estimation using Markov random fields (DEUM). DEUM is a subclass of estimation of distribution algorithms (EDAs) where interaction between solution variables is represented as an undirected graph and the joint probability of a solution is factorized as a Gibbs distribution derived from the structure of the graph. The focus of this paper will be on describing the three main characteristics of DEUM framework, which distinguishes it from the traditional EDA. They are: 1) use of MRF models, 2) fitness modeling approach to estimating the parameter of the model and 3) Monte Carlo approach to sampling from the model.
Similar content being viewed by others
References
J. H. Holland. Adaptation in Natural and Artificial Systems, University of Michigan Press, Ann Arbor, MI, 1975.
I. Rechenberg. Evolutionstrategie-optimierung Technischer Systeme nach Prinzipien der biologischen Evolution, Formman-Holzboog, Stuttgart, 1973.
L. J. Fogel. Autonomous Automata. Industrial Research, vol. 4, no. 1, pp. 14–19, 1962.
H. Mühlenbein, G. Paaß. From Recombination of Genes to the Estimation of Distributions: I. Binary Parameters. In Proceedings of the 4th International Conference on Parallel Problem Solving from Nature, Berlin, pp. 178–187, 1996.
P. Larrañaga, J. A. Lozano. Estimation of Distribution Algorithms: a New Tool for Evolutionary Computation, Kluwer Academic Publishers, Boston, 2002.
M. Pelikan, D. E. Goldberg. Hierarchical BOA Solves Ising Spin Glasses and MAXSAT. In Proceedings of the Genetic and Evolutionary Computation Conference, Springer Verlag, Chicago, IL, pp. 1275–1286, 2003.
J. A. Lozano, R. Sagarna, P. Larrañaga. Solving Job Scheduling with Estimation of Distribution Algorithms. In Estimation of Distribution Algorithms. A New Tool for Evolutionary Computation, P. Larrañaga and J. A. Lozano, (eds.), Kluwer Academis Publishers, Norwell, MA, pp. 231–242, 2001.
Q. Zhang, J. Sun, E. Tsang, An Evolutionary Algorithm with Guided Mutation for the Maximum Clique Problem. IEEE Transactions on Evolutionary Computation, vol. 9, no. 2, pp. 192–200, 2005.
S. K. Shakya, J. A. W. McCall, D. F. Brown. Updating the Probability Vector Using MRF Technique for a Univariate EDA. In Proceedings of the Second Starting AI Researchers’ Symposium, Frontiers in Artificial Intelligence and Applications, IOS press, Valencia, Spain, vol. 109, pp. 15–25, 2004.
S. Shakya. DEUM: A Framework for an Estimation of Distribution Algorithm Based on Markov Random Fields. Ph.D. dissertation. The Robert Gordon University, Aberdeen, UK, April 2006.
R. Santana. Estimation of Distribution Algorithms with Kikuchi Approximation. Evolutonary Computation, vol. 13, no. 1, pp. 67–98, 2005.
R. Kikuchi. A Theory of Cooperative Phenomena. Physical Review, vol. 81, no. 6, pp. 988–1003, 1951.
P. Larrañaga, R. Etxeberria, J. A. Lozano, B. Sierra, I. Inza, J. M. Peña. A Review of the Cooperation between Evolutionary Computation and Probabilistic Graphical Models. In Proceedings of the Second Symposium on Artificial Intelligence, La Habana, pp. 314–324, 1999.
M. I. Jordan, editor. Learning in Graphical Models, NATO Science Series. Kluwer Academic Publishers, Dordrecht, 1998.
J. Besag. Spatial Interaction and the Statistical Analysis of Lattice Systems. Journal of the Royal Statistical Society, vol. 36, no. 2, pp. 192–236, 1974.
S. Z. Li. Markov Random Field Modeling in Computer Vision, Springer-Verlag, London, UK, 1995.
K. P. Murphy. Dynamic Bayesian Networks: Representation, Inference and Learning. Ph.D. dissertation, University of California, Berkeley, 2002.
J. M. Hammersley, P. Clifford. Markov Fields on Finite Graphs and Lattices, Unpublished Manuscrpt, 1971.
D. F. Brown, A. B. Garmendia-Doval, J. A. W. McCall. Markov Random Field Modelling of Royal Road Genetic Algorithms. Lecture Notes in Computer Science, Springer, January, vol. 2310, pp. 65–78, 2002.
R. Kindermann, J. L. Snell. Markov Random Fields and Their Applications, American Mathematical Society, Providence, RI 02904-2294 USA, 1980.
W. H. Press, S. A. Teukolsky, W. T. Vetterling, B. P. Flannery. Numerical Recipes in C: The Art of Scientific Computing, 2nd edition, Cambridge University Press, Cambridge, UK, 1992.
S. Shakya, J. McCall, D. Brown. Estimating the Distribution in an EDA. In Proceedings of the International Conference on Adaptive and Natural Computing Algorithms, Coimbra, Portugal, pp. 202–205, 2005.
S. Shakya, J. McCall, Brown D. Using a Markov Network Model in a Univariate EDA: An Emperical Cost-Benefit Analysis. In Proceedings of Genetic and Evolutionary Computation Conference, Washington D.C., USA, pp. 727–734, 2005.
S. K. Shakya, J. A. W. McCall, D. F. Brown. Incorporating a Metropolis Method in a Distribution Estimation Using Markov Random Field Algorithm. In Proceedings of IEEE Congress on Evolutionary Computation. Edinburgh, UK, vol. 3, pp. 2576–2583, 2005.
S. K. Shakya, J. A. W. McCall, D. F. Brown. Solving the Ising Spin Glass Problem Using a Bivariate EDA Based on Markov Random Fields. In Proceedings of IEEE Congress on Evolutionary Computation, Vancouver, Canada, pp. 908–915, 2006.
N. Metropolis, A. W. Rosenbluth, M. N. Rosenbluth, A. H. Teller, E Teller. Equations of State Calculations by Fast Computational Machine. Journal of Chemical Physics, vol. 21, no. 6, pp. 1087–1091, 1953.
S. Geman, D. Geman. Stochastic Relaxation, Gibbs Distributions and the Bayesian Restoration of Images. Readings in Computer Vision: Issues, Problems, Principles, and Paradigms, M. A. Fischler, O. Firschein, (eds.), Kaufmann, Los Altos, CA., pp. 564–584, 1987.
M. Pelikan. Bayesian Optimization Algorithm: From Single Level to Hierarchy. Ph.D. dissertation, University of Illinois at Urbana-Champaign, Urbana, IL, 2002. Also IlliGAL Report No. 2002023.
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.
H. H. Hoss, T. Stutzle. Towards a Characterization of the Behaviour of Stochastic Local Search Algorithms for SAT. Artificial Intelligence, vol. 112, no. 1–2, pp. 213–232, 1999.
R. Santana. Probabilistic Modeling Based on Undirected Graphs in Estimation Distribution Algorithms. Ph.D. dissertation, Institute of Cybernetics, Mathematics and Physics, Havana, Cuba, 2003.
Author information
Authors and Affiliations
Corresponding author
Additional information
Siddhartha K. Shakya received his B.E. and M.E. degree in computer engineering from the Vladimir State University. Russia, in 1998 and 1999 respectively, and M.Sc. degree in intelligent systems from the University of Sussex. UK in 2002. He received his Ph.D. degree in computer science from the Robert Gordon University, Aberdeen, UK in 2006. Currently, he is a research fellow at the Intelligent Systems Research Center within BT research. Ipswich, UK. Prior to joining BT, he was a research assistant in School of Computing at the Robert Gordon University.
His research interests include theoretical and practical aspects of optimisation algorithms, particularly, evolutionary algorithms and estimation of distribution algorithms, probabilistic graphical models and Markov random fields, graph-drawing algorithms, machine learning techniques, bio-informatics and operational research techniques, such as revenue management and dynamic pricing. He is a member of IEEE and ACM societies in computational intelligence.
John McCall studied Pure Mathematics at the University of Aberdeen, where he completed a Ph.D. on the Stable Homotopy of Lie Groups in 1991. Since then his research interests have evolved through modeling and optimization to evolutionary algorithms and, latterly, strong focus on EDAs. He is particularly interested in Markov random field models and their potential to improve heuristic search by modeling fitness from samples of evaluated solutions. He also has a strong and abiding interest in applications of evolutionary algorithms to medical treatment optimization, in particular cancer chemotherapy, and has published extensively in this area. Other interests include particle swarms, ant colonies and artificial immune systems.
McCall is currently a senior lecturer in computing at Robert Gordon University where he leads the Computational Intelligence Research Group.
Rights and permissions
About this article
Cite this article
Shakya, S., McCall, J. Optimization by estimation of distribution with DEUM framework based on Markov random fields. Int J Automat Comput 4, 262–272 (2007). https://doi.org/10.1007/s11633-007-0262-6
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/s11633-007-0262-6