Inference in Multiply Sectioned Bayesian Networks with Extended Shafer-Shenoy and Lazy Propagation

As Bayesian networks are applied to larger and more complex problem domains, search for flexible modeling and more efficient in­ ference methods is an ongoing effort. Mul­ tiply sectioned Bayesian networks (MSBNs) extend the HUGIN inference for Bayesian networks into a coherent framework for flexible modeling and distributed inference. Lazy propagation extends the Shafer-Shenoy and HUG IN inference methods with reduced space complexity. We apply the Shafer-Shenoy and lazy propa­ gation to inference in MSBNs. The combina­ tion of the MSBN framework and lazy propa­ gation provides a better framework for mod­ eling and inference in very large domains. It retains the modeling flexibility of MSBNs and reduces the runtime space complexity, allow­ ing exact inference in much larger domains given the same computational resources.