Automated Refinement of Bayes Networks' Parameters based on Test Ordering Constraints

In this paper, we derive a method to refine a Bayes network diagnostic model by exploiting constraints implied by expert decisions on test ordering. At each step, the expert executes an evidence gathering test, which suggests the test's relative diagnostic value. We demonstrate that consistency with an expert's test selection leads to non-convex constraints on the model parameters. We incorporate these constraints by augmenting the network with nodes that represent the constraint likelihoods. Gibbs sampling, stochastic hill climbing and greedy search algorithms are proposed to find a MAP estimate that takes into account test ordering constraints and any data available. We demonstrate our approach on diagnostic sessions from a manufacturing scenario.

[1]  David Heckerman,et al.  Causal independence for probability assessment and inference using Bayesian networks , 1996, IEEE Trans. Syst. Man Cybern. Part A.

[2]  Ronald A. Howard,et al.  Information Value Theory , 1966, IEEE Trans. Syst. Sci. Cybern..

[3]  Qiang Ji,et al.  Learning Bayesian network parameters under incomplete data with domain knowledge , 2009, Pattern Recognit..

[4]  Anthony Jameson,et al.  Exploiting Qualitative Knowledge in the Learning of Conditional Probabilities of Bayesian Networks , 2000, UAI.

[5]  A. J. Feelders A new parameter Learning Method for Bayesian Networks with Qualitative Influences , 2007, UAI.

[6]  Qiang Ji,et al.  Improving Bayesian Network parameter learning using constraints , 2008, 2008 19th International Conference on Pattern Recognition.

[7]  David Heckerman,et al.  Decision-theoretic troubleshooting , 1995, CACM.

[8]  Ryszard S. Michalski,et al.  A theory and methodology of inductive learning , 1993 .

[9]  Tom M. Mitchell,et al.  Bayesian Network Learning with Parameter Constraints , 2006, J. Mach. Learn. Res..

[10]  Andrew W. Moore,et al.  Fast Information Value for Graphical Models , 2005, NIPS.

[11]  Pascal Poupart,et al.  Evaluation Results for a Query-Based Diagnostics Application , 2010 .

[12]  Marek J. Druzdzel,et al.  Elicitation of Probabilities for Belief Networks: Combining Qualitative and Quantitative Information , 1995, UAI.

[13]  Yi Mao,et al.  Domain Knowledge Uncertainty and Probabilistic Parameter Constraints , 2009, UAI.

[14]  Ross D. Shachter,et al.  Learning From What You Don't Observe , 1998, UAI.

[15]  Michael P. Wellman Fundamental Concepts of Qualitative Probabilistic Networks , 1990, Artif. Intell..

[16]  Pedro Gil,et al.  A procedure to test the suitability of a factor for stratification in estimating diversity , 1991 .

[17]  Dan Klein,et al.  Learning from measurements in exponential families , 2009, ICML '09.

[18]  Thomas G. Dietterich,et al.  Learning from Sparse Data by Exploiting Monotonicity Constraints , 2005, UAI.