Abstract
During past few years, a variety of methods have been developed for learning probabilistic networks from data, among which the heuristic single link forward or backward searches are widely adopted to reduce the search space. A major drawback of these search heuristics is that they can not guarantee to converge to the right networks even if a sufficiently large data set is available. This motivates us to explore a new algorithm that will not sutler from this problem. In this paper, we first identify an asymptotic property of different score metrics, based on which we then present a hybrid learning method that can be proved to be asymptotically convergent. We show that the algorithm, when employing the information criterion mid the I3ayesian metric, guarantee to converge in a very general way and is computationally feasible. Evaluation of the algorithm with simulated data is given to demonstrate the capability of the algorithm.
This is a preview of subscription content, log in via an institution.
Preview
Unable to display preview. Download preview PDF.
References
Cooper, G. and Herskovits, E, “A Bayesian method for the induction of probabilistic networks from data”, Machine Learning, 9, pp. 309–374, 1992.
Herskovits, E.H. and Cooper, G.F., Kutato, “An entropy-driven system for the construction of probabilistic expert systems from databases”, Proc. of the Conf on Uncertainty in Artificial Intelligence, pp. 54–62, Cambridge, MA, 1990.
Lam, W. And Bacchus, F., “Learning Bayesian belief networks, an approach based on the MDL principle”, Computational Intelligence, 10, pp. 269–293, 1994.
Buntinc, W, “Theory refinement on Bayesian networks”, Proceedings of Seventh Conference on Uncertainty in Artificial Intelligence, Los Angeles, CA, pp. 652–660, Morgan Kaufmann, 1991.
Heckcrman, D., Geiger, D., and Chickering, D. “Learning Bayesian networks: The combination of knowledge and statistical data”. Machine Learning, 20, pp. 197–234, 1995.
Xiang, Y., Wong, S.K.M., and Cercone, N., “Critical remarks on single link search in learning belief networks”, Proc. of the Conference on Uncertainly in Artificial Intelligence, pp. 564–571, 1996.
Chang, K.C. and Liu, J., “Efficient algorithms for learning Bayesian networks”, 1996 IEEE International Conference on Systems, Man, and Cybernetics, Vol. 2, pp. 1274–1279, 1996..
Pearl, J, Probabilistic reasoning in intelligent systems. San Mateo, CA, Morgan Kaufmann, 1988.
Liu, J., Model learning with probabilistic networks, a Ph.D. dissertation, School of Information Technology and Engineering, George Mason University, Fairfax, VA, 1997.
Beinlich, I.A., Suermondt, H.J., Chavez, R.M., & Cooper, G.F, “The ALARM monitoring system: A case study with two probabilistic inference techniques for belief networks”, Proceedings of the Second European Conference on Artificial Intelligence in Medicine, pp. 247–256, London, England, 1989.
Bouckacrt, R. R, Bayesian belief networks: from construction to inference, Ph.D. dissertation, University Utrecht, Dutch, 1995.
Chang, K.C. and Fung, R., “Symbolic probabilistic inference with both discrete and continuous variables”, IEEE Transaction on Systems, Man, and Cybernetics, Vol. 25, No. 6, pp. 910–916, 1995.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Liu, J., Chang, KC., Zhou, J. (1998). A hybrid convergent method for learning probabilistic networks. In: Mercer, R.E., Neufeld, E. (eds) Advances in Artificial Intelligence. Canadian AI 1998. Lecture Notes in Computer Science, vol 1418. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-64575-6_66
Download citation
DOI: https://doi.org/10.1007/3-540-64575-6_66
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-64575-7
Online ISBN: 978-3-540-69349-9
eBook Packages: Springer Book Archive