Approximate Inference Algorithms for Hybrid Bayesian Networks with Discrete Constraints

In this paper, we consider Hybrid Mixed Networks (HMN) which are Hybrid\nBayesian Networks that allow discrete deterministic information to be modeled\nexplicitly in the form of constraints. We present two approximate inference\nalgorithms for HMNs that integrate and adjust well known algorithmic principles\nsuch as Generalized Belief Propagation, Rao-Blackwellised Importance Sampling\nand Constraint Propagation to address the complexity of modeling and reasoning\nin HMNs. We demonstrate the performance of our approximate inference algorithms\non randomly generated HMNs.\n