Hypothesis Management in Situation-Specific Network Construction

Citation: Laskey, K.B., S.M. Mahoney, and E.J. Wright, "Hypothesis Management in Situation-Specific Network Construction", Uncertainty in Artificial Intelligence: Proceedings of the Seventeenth Conference, J. Breese and D. Koller (eds), San Mateo, CA: Morgan Kaufmann.


This paper considers the problem of knowledgebased model construction in the presence of uncertainty about the association of domain entities to random variables. Multi-entity Bayesian networks (MEBNs) are defined as a representation for knowledge in domains characterized by uncertainty in the number of relevant entities, their interrelationships, and their association with observables. An MEBN implicitly specifies a probability distribution in terms of a hierarchically structured collection of Bayesian network fragments that together encode a joint probability distribution over arbitrarily many interrelated hypotheses. Although a finite query-complete model can always be constructed, association uncertainty typically makes exact model construction and evaluation intractable. The objective of hypothesis management is to balance tractability against accuracy. We describe an approach to hypothesis management, present an application to the problem of military situation awareness, and compare our approach to related work in the tracking and fusion literature.

For more information, contact IET