Exploiting Dynamic Independence in a Static Conditioning Graph
A conditioning graph (CG) is a graphical structure that attempt to minimize the implementation overhead of computing probabilities in belief networks. A conditioning graph recursively factorizes the network, but restricting each decomposition to a single node allows us to store the structure with minimal overhead, and compute with a simple algorithm. This paper extends conditioning graphs with optimizations that effectively reduce the height of the CG, thus reducing time complexity exponentially, while increasing the storage requirements by only a constant factor. We conclude that CGs are frequently as efficient as any other exact inference method, with the advantage of being vastly superior to VE and JT in terms of space complexity, and far simpler to implement.
KeywordsBayesian Network Leaf Node Internal Node Conditional Probability Table Irrelevant Variable
Unable to display preview. Download preview PDF.
- 1.Allen, D., Darwiche, A.: Optimal time–space tradeoff in probabilistic inference. In: Proceedings of the 18th International Joint Conference on Artificial Intelligence (IJCAI-2003), pp. 969–975 (2003)Google Scholar
- 2.Darwiche, A.: Recursive Conditioning: Any-space conditioning algorithm with treewidth-bounded complexity. Artificial Intelligence, 5–41 (2000)Google Scholar
- 3.Darwiche, A., Provan, G.: Query dags: A practical paradigm for implementing belief network inference. In: Proceedings of the 12th Annual Conference on Uncertainty in Artificial Intelligence, pp. 203–210 (1996)Google Scholar
- 6.Grant, K., Horsch, M.: Conditioning Graphs: Practical Structures for Inference in Bayesian Networks. In: Proceedings of the The 18th Australian Joint Conference on Artificial Intelligence, pp. 49–59 (2005)Google Scholar
- 7.Grant, K., Horsch, M.: Methods for Constructing Balanced Elimination Trees and Other Recursive Decompositions. In: Proceedings of the the 19th International Florida Artificial Intelligence Research Society Conference (to appear, 2006)Google Scholar
- 8.Kjaerulff, U.: Triangulation of graphs - algorithms giving small total state space. Technical report, Dept. of Mathematics and Computer Science, Strandvejan, DK 9000 Aalborg, Denmark (1990)Google Scholar
- 11.Pearl, J.: Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann Publishers Inc., San Francisco (1988)Google Scholar
- 14.Shachter, R.D.: Bayes-ball: The rational pastime (for determining irrelevance and requisite information in belief networks and influence diagrams). In: Proceedings of the 14th Annual Conference on Uncertainty in Artificial Intelligence, pp. 480–487 (1998)Google Scholar
- 15.Zhang, N., Poole, D.: A Simple Approach to Bayesian Network Computations. In: Proc. of the 10th Canadian Conference on Artificial Intelligence, pp. 171–178 (1994)Google Scholar