Approximate Inference Algorithms for Hybrid Bayesian Networks with Discrete Constraints
This work addresses the complexity of modeling and reasoning in hybrid Bayesian networks with constraints, which is an incremental improvement for researchers in probabilistic graphical models.
The paper tackles the problem of performing approximate inference in Hybrid Mixed Networks, which are Hybrid Bayesian Networks with discrete constraints, by presenting two algorithms that integrate Generalized Belief Propagation, Rao-Blackwellised Importance Sampling, and Constraint Propagation, and demonstrates their performance on randomly generated networks.
In this paper, we consider Hybrid Mixed Networks (HMN) which are Hybrid Bayesian Networks that allow discrete deterministic information to be modeled explicitly in the form of constraints. We present two approximate inference algorithms for HMNs that integrate and adjust well known algorithmic principles such as Generalized Belief Propagation, Rao-Blackwellised Importance Sampling and Constraint Propagation to address the complexity of modeling and reasoning in HMNs. We demonstrate the performance of our approximate inference algorithms on randomly generated HMNs.