Advertisement

The Gaussian Polytree EDA with Copula Functions and Mutations

  • Ignacio Segovia Domínguez
  • Arturo Hernández Aguirre
  • Enrique Villa Diharce
Part of the Studies in Computational Intelligence book series (SCI, volume 447)

Abstract

This chapter introduces the Gaussian Poly-Tree Estimation Distribution Algorithm, and two extensions: i) with Gaussian copula functions, and ii) with local optimizers. The new construction and simulation algorithms, and its application to estimation of distribution algorithms with continuous Gaussian variables are also introduced. The algorithm for the construction of the structure and for edge orientation is based on information theoretic concepts such as mutual information and conditional mutual information. The three models are tested on a benchmark of 20 unimodal and multimodal functions. The version with copula function and mutations excels in most problems achieving near optimal success rate.

Keywords

Mutual Information Multivariate Normal Distribution Joint Probability Density Function Copula Function Copula Modeling 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Acid, S., de Campos, L.M.: Approximations of Causal Networks by Polytrees: an Empirical Study. In: Bouchon-Meunier, B., Yager, R.R., Zadeh, L.A. (eds.) IPMU 1994. LNCS, vol. 945, pp. 149–158. Springer, Heidelberg (1995)CrossRefGoogle Scholar
  2. 2.
    Bacigál, T., Komorníková, M.: Fitting archimedean copulas to bivariate geodetic data. In: Rizzi, A., Vichi, M. (eds.) Compstat 2006 Proceedings in Computational Statistics, pp. 649–656. Physica-Verlag HD, Heidelberg (2006)Google Scholar
  3. 3.
    Bosman, P.A.N., Grahl, J., Thierens, D.: Enhancing the performance of maximum-likelihood gaussian edas using anticipated mean shift. In: Proceedings of BNAIC 2008, the Twentieth Belgian-Dutch Artificial Intelligence Conference, pp. 285–286. BNVKI (2008)Google Scholar
  4. 4.
    Castillo, E., Gutiérrez, J.M., Hadi, A.S.: Expert Systems and Probabilistic Network Models. Springer (1997)Google Scholar
  5. 5.
    Cherubini, U., Luciano, E., Vecchiato, W.: Copula Methods in Finance. Wiley, Chichester (2004)zbMATHGoogle Scholar
  6. 6.
    Chikering, D.M.: Learning bayesian networks is np-complete. In: Learning from Data: Artificial Intelligence and Statistics V, pp. 121–130. Springer (1996)Google Scholar
  7. 7.
    Chow, C.K., Liu, C.N.: Approximating discrete probability distributions with dependence trees. IEEE Transactions on Information Theory IT-14(3), 462–467 (1968)Google Scholar
  8. 8.
    Darwiche, A.: Modeling and Reasoning with Bayesian Networks. Cambridge University Press (2009)Google Scholar
  9. 9.
    Dasgupta, S.: Learning polytrees. In: Proceedings of the Fifteenth Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI 1999), pp. 134–14. Morgan Kaufmann, San Francisco (1999)Google Scholar
  10. 10.
    De-Waal, D.J., Van-Gelder, P.H.A.J.M.: Modelling of extreme wave heights and periods through copulas. Extremes 8(4), 345–356 (2005)MathSciNetzbMATHCrossRefGoogle Scholar
  11. 11.
    Edwards, D.: Introduction to Graphical Modelling. Springer, Berlin (1995)zbMATHGoogle Scholar
  12. 12.
    Genest, C., Favre, A.C.: Everything you always wanted to know about copula modeling but were afraid to ask. Journal of Hydrologic Engineering 12(4), 347–368 (2007)CrossRefGoogle Scholar
  13. 13.
    Bosman, P.A.N., Grahl, J., Rothlauf, F.: The correlation-triggered adaptive variance scaling idea. In: GECCO 2006, Proceedings of the 8th Annual Conference on Genetic and Evolutionary Computation, pp. 397–404. ACM (2006)Google Scholar
  14. 14.
    Joe, H.: Multivariate models and dependence concepts. Chapman and Hall, London (1997)zbMATHGoogle Scholar
  15. 15.
    Lozano, I.I.J.A., Larrañaga, P., Bengoetxea, E., Larrañaga, P., Bengoetxea, E.: Towards a New Evolutionary Computation: Advances on Estimation of Distributions Algorithms. STUDFUZZ. Springer (2006)Google Scholar
  16. 16.
    Lauritzen, S.L.: Graphical models. Clarendon Press (1996)Google Scholar
  17. 17.
    Oommen, B.J., Ouerd, M., Matwin, S.: A formal approach to using data distributions for building causal polytree structures. Information Sciences, an International Journal 168, 111–132 (2004)MathSciNetzbMATHGoogle Scholar
  18. 18.
    Monjardin, P.E.: Análisis de dependencia en tiempo de falla. Master’s thesis, Centro de Investigación en Matemáticas, Guanajuato, México (December 2007) (in Spanish)Google Scholar
  19. 19.
    Neapolitan, R.E.: Learning Bayesian Networks. Prentice Hall series in Artificial Intelligence (2004)Google Scholar
  20. 20.
    Nelsen, R.B.: An Introduction to Copulas, 2nd edn. Springer Series in Statistics. Springer (2006)Google Scholar
  21. 21.
    Ortiz, M.S.: Un estudio sobre los Algoritmos Evolutivos con Estimacion de Distribuciones basados en poliarboles y su costo de evaluacion. PhD thesis. Instituto de Cibernetica, Matematica y Fisica, La Habana, Cuba (2003)Google Scholar
  22. 22.
    Ouerd, M.: Learning in Belief Networks and its Application to Distributed Databases. PhD thesis. University of Ottawa, Ottawa, Ontario, Canada (2000)Google Scholar
  23. 23.
    Pearl, J.: Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann Publishers Inc., San Francisco (1988)Google Scholar
  24. 24.
    de Campos, L.M., Molina, J.M.R.: Using bayesian algorithms for learning causal networks in classification problems. In: Proceedings of the Fourth International Conference of Information Processing and Management of Uncertainty in Knowledge-Based Systems (IPMU), pp. 395–398 (1993)Google Scholar
  25. 25.
    Rebane, G., Pearl, J.: The recovery of causal poly-trees from statistical data. In: Proceedings, 3rd Workshop on Uncertainty in AI, Seattle, WA, pp. 222–228 (1987)Google Scholar
  26. 26.
    Schölzel, C., Friederichs, P.: Multivariate non-normally distributed random variables in climate research – introduction to the copula approach. Nonlinear Processes in Geophysics 15(5), 761–772 (2008)CrossRefGoogle Scholar
  27. 27.
    Shachter, R., Kenley, C.: Gaussian influence diagrams. Management Science 35(5), 527–550 (1989)CrossRefGoogle Scholar
  28. 28.
    Sklar, A.: Fonctions de répartition à n dimensions et leurs marges. Publications de l’Institut de Statistique de l’Université de Paris 8, 229–231 (1959)MathSciNetGoogle Scholar
  29. 29.
    Trivedi, P.K., Zimmer, D.M.: Copula Modeling: An Introduction for Practitioners. Foundations and Trends® in Econometrics, vol. 1. Now Publishers (2007)Google Scholar
  30. 30.
    Whittaker, J.: Graphical Models in Applied Multivariate Statistics. John Wiley & Sons (1990)Google Scholar

Copyright information

© Springer Berlin Heidelberg 2013

Authors and Affiliations

  • Ignacio Segovia Domínguez
    • 1
  • Arturo Hernández Aguirre
    • 1
  • Enrique Villa Diharce
    • 1
  1. 1.Center for Research in MathematicsGuanajuatoMéxico

Personalised recommendations