Optimization of Inter-Subnet Belief Updating in Multiply Sectioned Bayesian Networks
This work addresses a computational bottleneck in MSBNs, which are used for diagnosis in natural and artificial systems, but it is incremental as it builds on existing methods.
The paper tackles the problem of inefficient belief propagation between subnets in Multiply Sectioned Bayesian Networks (MSBNs) by analyzing and optimizing the UpdateBelief operation, resulting in two new versions that reduce computation time, with one achieving optimal efficiency through a graph-theoretic solution.
Recent developments show that Multiply Sectioned Bayesian Networks (MSBNs) can be used for diagnosis of natural systems as well as for model-based diagnosis of artificial systems. They can be applied to single-agent oriented reasoning systems as well as multi-agent distributed probabilistic reasoning systems. Belief propagation between a pair of subnets plays a central role in maintenance of global consistency in a MSBN. This paper studies the operation UpdateBelief, presented originally with MSBNs, for inter-subnet propagation. We analyze how the operation achieves its intended functionality, which provides hints as for how its efficiency can be improved. We then define two new versions of UpdateBelief that reduce the computation time for inter-subnet propagation. One of them is optimal in the sense that the minimum amount of computation for coordinating multi-linkage belief propagation is required. The optimization problem is solved through the solution of a graph-theoretic problem: the minimum weight open tour in a tree.