Bayesian Error-Bars for Belief Net Inference

A Bayesian Belief Network (BN) is a model of a joint distribution over a finite set of variables, with a DAG structure to represent the immediate dependencies between the variables, and a set of parameters (aka CPTables) to represent the local conditional probabilities of a node, given each assignment to its parents. In many situations, the parameters are themselves treated as random variables -- reflecting the uncertainty remaining after drawing on knowledge of domain experts and/or observing data generated by the network. A distribution over the CPtable parameters induces a distribution for the response the BN will return to any "What is Pr{H/E}?" query. This paper investigates the distribution of this response, shows that it is asymptotically normal, and derives expressions for its mean and asymptotic variance. We show that this computation has the same complexity as simply computing the (mean value of the) response -- i.e., O(n exp(w)), where n is the number of variables and w is the effective tree width. We also provide empirical evidence showing that the error-bars computed from our estimates are fairly accurate in practice, over a wide range of belief net structures and queries.

[1]  Yoshihisa Akimoto A NOTE ON UNIFORM ASYMPTOTIC NORMALITY OF DIRICHLET DISTRIBUTION , 1996 .

[2]  Richard E. Neapolitan,et al.  An Implementation of a Method for Computing the Uncertainty in Inferred Probabilities in Belief Networks , 1993, UAI.

[3]  Adnan Darwiche,et al.  A differential approach to inference in Bayesian networks , 2000, JACM.

[4]  Craig Boutilier,et al.  Context-Specific Independence in Bayesian Networks , 1996, UAI.

[5]  Stephen E. Fienberg,et al.  Discrete Multivariate Analysis: Theory and Practice , 1976 .

[6]  Dale Schuurmans,et al.  Learning Bayesian Nets that Perform Well , 1997, UAI.

[7]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems - networks of plausible inference , 1991, Morgan Kaufmann series in representation and reasoning.

[8]  Prakash P. Shenoy,et al.  A Comparison of Lauritzen-Spiegelhalter, Hugin, and Shenoy-Shafer Architectures for Computing Marginals of Probability Distributions , 1998, UAI.

[9]  Ron Musick,et al.  Minimal Assumption Distribution Propagation in Belief Networks , 1993, UAI.

[10]  Gernot D. Kleiter,et al.  Propagating Imprecise Probabilities in Bayesian Networks , 1996, Artif. Intell..

[11]  Kathryn B. Laskey Sensitivity analysis for probability assessments in Bayesian networks , 1995, IEEE Trans. Syst. Man Cybern..

[12]  Gregory F. Cooper,et al.  The ALARM Monitoring System: A Case Study with two Probabilistic Inference Techniques for Belief Networks , 1989, AIME.

[13]  Russell Greiner,et al.  Efficient reasoning , 2001, CSUR.

[14]  Gregory F. Cooper,et al.  The Computational Complexity of Probabilistic Inference Using Bayesian Belief Networks , 1990, Artif. Intell..

[15]  Rina Dechter,et al.  Bucket elimination: A unifying framework for probabilistic inference , 1996, UAI.

[16]  Wc Taylor,et al.  Special Issue , 2000, International Journal of Recent Technology and Engineering.

[17]  G F Cooper,et al.  Algorithms for Bayesian Belief-Network Precomputation , 1991, Methods of Information in Medicine.