Statistics and Computing

, Volume 8, Issue 2, pp 159–173 | Cite as

An efficient algorithm for finding the M most probable configurationsin probabilistic expert systems



A probabilistic expert system provides a graphical representation of a joint probability distribution which enables local computations of probabilities. Dawid (1992) provided a ‘flow- propagation’ algorithm for finding the most probable configuration of the joint distribution in such a system. This paper analyses that algorithm in detail, and shows how it can be combined with a clever partitioning scheme to formulate an efficient method for finding the M most probable configurations. The algorithm is a divide and conquer technique, that iteratively identifies the M most probable configurations.

Bayesian network belief revision charge conditional independence divide-and-conquer evidence flow junction tree marginalization maximization most probable explanation potential function propagation 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Andersen, S. K., Olesen, K. G., Jensen, F. V. and Jensen, F. (1990) HUGIN — a shell for building Bayesian belief universes for expert systems. In G. Shafer and J. Pearl (eds) Readings in Uncertainty, pp. 332–337. San Francisco: Morgan Kaufmann.Google Scholar
  2. Cowell, R. G. (1992) BAIES — a probabilistic expert system shell with qualitative and quantitative learning. In J. M. Bernardo, J. O. Berger, A. P. Dawid and A. F. M. Smith (eds) Bayesian Statistics 4th edn, pp. 595–600. Oxford: Clarendon Press.Google Scholar
  3. Dawid, A. P. (1992) Applications of a general propagation algorithm for probabilistic expert systems. Statistics and Computing, 2, 25–36.Google Scholar
  4. Jensen, F. V. (1988) Junction trees and decomposable hypergraphs. Judex Research Report, Judex Datasystemer A/S, Aalborg, Denmark F. V. (1990) Calculation in HUGIN of probabilities.Google Scholar
  5. Jensen, F. V., Olesen, K. G. and Andersen, S. K. (1990) An algebra of Bayesian belief universes for knowledge-based systems. Networks, 20, 637–59.Google Scholar
  6. Jensen, F. V. (1996) Introduction to Bayesian Networks. London: UCL Press.Google Scholar
  7. Lauritzen, S. L., Speed, T. P. and Vijayan, K. (1984) Decomposable graphs and hypergraphs. Journal of the Australian Mathematical Society, Series A, 36, 12–29.Google Scholar
  8. Lauritzen, S. L. and Spiegelhalter, D. J. (1988) Local computations with probabilities on graphical structures and their application to expert systems (with discussion). Journal of the Royal Statistical Society Series B, 50, 157–224.Google Scholar
  9. Li, Z. and D'Ambrosio, B. (1993) An efficient approach for finding MAP assignments to belief networks. In Proceedings of the 9th Conference on Uncertainty in AI, pp. 342–9. Washington, Morgan Kaufmann.Google Scholar
  10. Nilsson, D. (1994) An algorithm for finding the M most probable configurations of discrete variables that are specified in probabilistic expert systems. MSc thesis, University of Copenhagen.Google Scholar
  11. Pearl, J. (1986) Fusion, propagation and structuring in belief networks. Artificial Intelligence, 29, 241–88.Google Scholar
  12. Pearl, J. (1988) Probabilistic Reasoning in Intelligence Systems. San Mateo, California: Morgan Kaufmann.Google Scholar
  13. Seroussi, B. and Golmard, J. L. (1994) An algorithm for finding the K most probable configurations in Bayesian networks. International Journal of Approximate Reasoning, 1, 205–33.Google Scholar
  14. Sy, B. K. (1993) A recurrence local computation approach towards ordering composite beliefs in Bayesian belief networks. International Journal of Approximate Reasoning, 8, 17–50.Google Scholar
  15. Xu, H. (1994) Computing marginals from the marginal representation in Markov Trees. Proceedings from the International Conference on Information Processing and Management of Uncertainty in Knowledge-based systems (IPMU), Paris, France, pp. 275–80. Paris, Cite Internationale Universitaire.Google Scholar

Copyright information

© Kluwer Academic Publishers 1998

Authors and Affiliations

    • 1
  1. 1.Department of Mathematics and Computer Science, Institute for Electronic SystemsAalborg UniversityAalborg ØstDenmark

Personalised recommendations