Interval Influence Diagrams

We describe a mechanism for performing probabilistic reasoning in influence diagrams using interval rather than point valued probabilities. We derive the procedures for node removal (corresponding to conditional expectation) and arc reversal (corresponding to Bayesian conditioning) in influence diagrams where lower bounds on probabilities are stored at each node. The resulting bounds for the transformed diagram are shown to be optimal within the class of constraints on probability distributions that can be expressed exclusively as lower bounds on the component probabilities of the diagram. Sequences of these operations can be performed to answer probabilistic queries with indeterminacies in the input and for performing sensitivity analysis on an influence diagram. The storage requirements and computational complexity of this approach are comparable to those for point-valued probabilistic inference mechanisms, making the approach attractive for performing sensitivity analysis and where probability information is not available. Limited empirical data on an implementation of the methodology are provided.

[1]  Benjamin N. Grosof.  An Inequality Paradigm for Probabilistic Knowledge: The Logic of Conditional Probability Intervals , 1985, UAI.

[2]  Scott M. Olmsted.  On representing and solving decision problems , 1983 .

[3]  Matthew L. Ginsberg,et al.  Does Probability Have a Place in Non-monotonic Reasoning? , 1985, IJCAI.

[4]  Ross D. Shachter.  Evaluating Influence Diagrams , 1986, Oper. Res..

[5]  Peter Haddawy.  Implementation of and Experiments with a Variable Precision Logic Inference System , 1986, AAAI.

[6]  Benjamin N. Grosof,et al.  An inequality paradigm for probabilistic knowledge the augmented logic of conditional probability intervals , 1985, UAI 1985.

[7]  Chelsea C. White,et al.  A Posteriori Representations Based on Linear Inequality Descriptions of a Priori and Conditional Probabilities , 1986, IEEE Transactions on Systems, Man, and Cybernetics.

[8]  H. Moskowitz,et al.  Robust interactive decision-analysis (RID): concepts, methodology, and system principles , 1989, [1989] Proceedings of the Twenty-Second Annual Hawaii International Conference on System Sciences. Volume III: Decision Support and Knowledge Based Systems Track.

[9]  Joseph B. Kadane,et al.  Rethinking the Foundations of Statistics: Decisions Without Ordering , 1990 .

[10]  Gregory F. Cooper,et al.  NESTOR: A Computer-Based Medical Diagnostic Aid That Integrates Causal and Probabilistic Knowledge. , 1984 .

[11]  J. Pearl.  On probability intervals , 1988, Int. J. Approx. Reason..

[12]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems , 1988 .

[13]  Henry E. Kyburg,et al.  Bayesian and Non-Bayesian Evidential Updating , 1987, Artificial Intelligence.

[14]  J. Kyburg.  Higher order probability and intervals , 1988 .

[15]  Isaac Levi,et al.  The Enterprise Of Knowledge , 1980 .

[16]  Henry E. Kyburg,et al.  Higher order probabilities and intervals , 1988, Int. J. Approx. Reason..

[17]  Paul Snow.  Bayesian Inference without Point Estimates , 1986, AAAI.