A join tree probability propagation architecture for semantic modeling
 C. J. Butz,
 H. Yao,
 S. Hua
 … show all 3 hide
Abstract
We propose the first join tree (JT) propagation architecture that labels the probability information passed between JT nodes in terms of conditional probability tables (CPTs) rather than potentials. By modeling the task of inference involving evidence, we can generate three work schedules that are more timeefficient for LAZY propagation. Our experimental results, involving five realworld or benchmark Bayesian networks (BNs), demonstrate a reasonable improvement over LAZY propagation. Our architecture also models inference not involving evidence. After the CPTs identified by our architecture have been physically constructed, we show that each JT node has a sound, local BN that preserves all conditional independencies of the original BN. Exploiting inference not involving evidence is used to develop an automated procedure for building multiply sectioned BNs. It also allows direct computation techniques to answer localized queries in local BNs, for which the empirical results on a realworld medical BN are promising. Screen shots of our implemented system demonstrate the improvements in semantic knowledge.
 Allen, D., & Darwiche, A. (2003). Optimal timespace tradeoff in probabilistic inference, In Proc. 18th international joint conference on artificial intelligence (pp. 969975). Acapulco, Mexico.
 Becker, A., Geiger, D. (2001) A sufficiently fast algorithm for finding close to optimal clique trees. Artificial Intelligence 125: pp. 317 CrossRef
 Castillo, E., Gutiérrez, J., Hadi, A. (1997) Expert systems and probabilistic network models. Springer, New York
 Consortium, E. (2002). Elvira: An environment for probabilistic graphical models. In Proceedings of the 1st European workshop on probabilistic graphical models (pp. 222–230). Cuenca, Espana.
 Cooper, G. F. (1990) The computational complexity of probabilistic inference using Bayesian belief networks. Artificial Intelligence 42: pp. 393405 CrossRef
 Cormen, T. H., Leiserson, C. E., Rivest, R. L., Stein, C. (2001) Introduction to algorithms. MIT, Toronto
 Cowell, R. G., Dawid, A. P., Lauritzen, S. L., Spiegelhalter, D. J. (1999) Probabilistic networks and expert systems. Springer, New York
 Dechter, R. (1996). Bucket elimination: A unifying framework for probabilistic inference. In Proc. 12th conference on uncertainty in artificial intelligence (pp. 211–219). Portland, OR.
 Hájek, P., Havránek, T., Jiroušek, R. (1992) Uncertain information processing in expert systems. CRC, Ann Arbor
 Jensen, F. V. (1996) An introduction to Bayesian networks. UCL, London
 Jensen, F. V., Lauritzen, S. L., Olesen, K. G. (1990) Bayesian updating in causal probabilistic networks by local computations. Computational Statistics Quarterly 4: pp. 269282
 Kjaerulff, U. (1990). Triangulation of graphs—algorithms giving small total state space, Research Report R9009. Dept. of Math. and Comp. Sci., Aalborg University, Denmark.
 Kozlov, A. V., Singh, J. P. (1999) Parallel implementations of probabilistic inference. IEEE Computer 29: pp. 3340
 Lauritzen, S. L., Spiegelhalter, D. J. (1988) Local computations with probabilities on graphical structures and their application to expert systems. Journal of the Royal Statistical Society Series B 50: pp. 157244
 Lauritzen, S. L., Dawid, A. P., Larsen, B. N., Leimer, H. G. (1990) Independence properties of directed Markov fields. Networks 20: pp. 491505 CrossRef
 Li, Z., D’Ambrosio, B. (1994) Efficient inference in Bayes networks as a combinatorial optimization problem. International Journal of Approximate Reasoning 11: pp. 5581 CrossRef
 Madsen, A. L., Jensen, F. V. (1999) LAZY propagation: A junction tree inference algorithm based on lazy evaluation. Artificial Intelligence 113: pp. 203245 CrossRef
 Madsen, A. L., & Jensen, F. V. (1999). Parallelization of inference in Bayesian networks. Research Report R995002, Dept. of Comp. Sci., Aalborg University, Denmark
 Neapolitan, R. E. (1990) Probabilistic reasoning in expert systems. Wiley, Toronto
 Olesen, K. G., Madsen, A. L. (2002) Maximal prime subgraph decomposition of Bayesian networks. IEEE Transactions on Systems, Man and Cybernetics B 32: pp. 2131 CrossRef
 Pearl, J. (1988) Probabilistic reasoning in intelligent systems: Networks of plausible inference. Morgan Kaufmann, San Francisco
 Schmidt, T., Shenoy, P. P. (1998) Some improvements to the ShenoyShafer and Hugin architectures for computing marginals. Artificial Intelligence 102: pp. 323333 CrossRef
 Shachter, R. (1986) Evaluating influence diagrams. Operational Research 34: pp. 871882
 Shafer, G., Shenoy, P. P. (1990) Probability propagation. Annals of Mathematics and Artificial Intelligence 2: pp. 327352 CrossRef
 Shafer, G. (1996) Probabilistic expert systems. SIAM, Philadelphia
 Wong, S. K. M., Butz, C. J., Wu, D. (2000) On the implication problem for probabilistic conditional independency. IEEE Transactions on Systems, Man and Cybernetics A 30: pp. 785805 CrossRef
 Xiang, Y. (1996) A probabilistic framework for cooperative multiagent distributed interpretation and optimization of communication. Artificial Intelligence 87: pp. 295342 CrossRef
 Xiang, Y. (2002) Probabilistic reasoning in multiagent systems: A graphical models approach. Cambridge University Press, New York
 Xiang, Y., & Jensen, F. V. (1999). Inference in multiply sectioned Bayesian networks with extended ShaferShenoy and Lazy propagation, In Proc. 15th conference on uncertainty in artificial intelligence (pp. 680–687). Stockholm, Sweden.
 Xiang, Y., Jensen, F. V., Chen, X. (2006) Inference in multiply sectioned Bayesian networks: Methods and performance comparison. IEEE Transactions on Systems, Man and Cybernetics B 36: pp. 546558 CrossRef
 Xiang, Y., Olesen, K. G., Jensen, F. V. (2000) Practical issues in modeling large diagnostic systems with multiply sectioned Bayesian networks. International Journal of Pattern Recognition and Artificial Intelligence 14: pp. 5971 CrossRef
 Xiang, Y., Pant, B., Eisen, A., Beddoes, M. P., Poole, D. (1993) Multiply sectioned Bayesian networks for neuromuscular diagnosis. Artificial Intelligence in Medicine 5: pp. 293314 CrossRef
 Xu, H. (1995) Computing marginals for arbitrary subsets from marginal representation in Markov trees. Artificial Intelligence 74: pp. 177189 CrossRef
 Yannakakis, M. (1981) Computing the minimal fillin is NPComplete. SIAM Journal on Algebraic and Discrete Methods 2: pp. 7779 CrossRef
 Zhang, N. L. (1998) Computational properties of two exact algorithms for Bayesian networks. Applied Intelligence 9: pp. 173184 CrossRef
 Title
 A join tree probability propagation architecture for semantic modeling
 Open Access
 Available under Open Access This content is freely available online to anyone, anywhere at any time.
 Journal

Journal of Intelligent Information Systems
Volume 33, Issue 2 , pp 145178
 Cover Date
 20091001
 DOI
 10.1007/s1084400800734
 Print ISSN
 09259902
 Online ISSN
 15737675
 Publisher
 Springer US
 Additional Links
 Topics
 Keywords

 Bayesian networks
 Join trees
 Probabilistic inference
 Conditional independence
 Industry Sectors
 Authors

 C. J. Butz ^{(1)}
 H. Yao ^{(1)}
 S. Hua ^{(1)}
 Author Affiliations

 1. Department of Computer Science, University of Regina, Regina, S4S 0A2, Canada