Title :
Some practical issues in inference in hybrid Bayesian networks with deterministic conditionals
Author :
Shenoy, Prakash P. ; Rumí, Rafael ; Salmerón, Antonio
Author_Institution :
Sch. of Bus., Univ. of Kansas, Lawrence, KS, USA
Abstract :
In this paper we analyze the use of hybrid Bayesian networks in domains that include deterministic conditionals for continuous variables. We show how exact inference can become infeasible even for small networks, due to the difficulty in handling functional relationships. We compare two strategies for carrying out the inference task, using mixtures of polynomials (MOPs) and mixtures of truncated exponentials (MTEs).
Keywords :
belief networks; deterministic algorithms; inference mechanisms; polynomials; continuous variables; deterministic conditionals; functional relationship handling; hybrid Bayesian networks; inference task; mixtures-of-polynomials; mixtures-of-truncated exponentials; Approximation methods; Bayesian methods; Hypercubes; Intelligent systems; Polynomials; Random variables; Stochastic processes; deterministic conditionals; hybrid Bayesian networks; mixtures of polynomials; mixtures of truncated exponentials;
Conference_Titel :
Intelligent Systems Design and Applications (ISDA), 2011 11th International Conference on
Conference_Location :
Cordoba
Print_ISBN :
978-1-4577-1676-8
DOI :
10.1109/ISDA.2011.6121722
