Previous |  Up |  Next


compositional models; marginalization; Bayesian network
Efficient computational algorithms are what made graphical Markov models so popular and successful. Similar algorithms can also be developed for computation with compositional models, which form an alternative to graphical Markov models. In this paper we present a theoretical basis as well as a scheme of an algorithm enabling computation of marginals for multidimensional distributions represented in the form of compositional models.
[1] Badsberg J. H.: An Environment for Graphical Models. Ph.D. Thesis, Aalborg University 1995.
[3] Jensen F. V.: Bayesian Networks and Decision Graphs. Springer Verlag, New York 2001 MR 1876880
[4] Jiroušek R.: Marginalization in composed probabilistic models. In: Proc. 16th Conf. Uncertainty in Artificial Intelligence UAI’00 (C. Boutilier and M. Goldszmidt, eds.), Morgan Kaufmann, San Francisco 2000, pp. 301–308
[5] Jiroušek R.: Decomposition of multidimensional distributions represented by perfect sequences. Ann. Math. and Artificial Intelligence 35 (2002), 215–226 DOI 10.1023/A:1014591402750 | MR 1899952 | Zbl 1004.60010
[6] Jiroušek R.: What is the difference between Bayesian networks and compositional models? In: Proc. 7th Czech-Japan Seminar on Data Analysis and Decision Making under Uncertainty (H. Noguchi, H. Ishii, M. Inuiguchi, eds.), Awaji Yumebutai ICC 2004, pp. 191–196
[7] Lauritzen S. L.: Graphical Models. Clarendon Press, Oxford 1996 MR 1419991 | Zbl 1055.62126
[8] Shachter R. D.: Evaluating influence diagrams. Oper. Res. 34 (1986), 871–890 DOI 10.1287/opre.34.6.871 | MR 0886655
[9] Shachter R. D.: Probabilistic inference and influence diagrams. Oper. Res. 36 (1988), 589–604 DOI 10.1287/opre.36.4.589 | Zbl 0651.90043
[10] Shafer G.: Probabilistic Expert Systems. SIAM, Philadelphia 1996 MR 1400892 | Zbl 0866.68108
Partner of
EuDML logo