A Variational Approximation for Bayesian Networks with Discrete and Continuous Latent Variables

We show how to use a variational approximation to the logistic function to perform approximate inference in Bayesian networks containing discrete nodes with continuous parents. Essentially, we convert the logistic function to a Gaussian, which facilitates exact inference, and then iteratively adjust the variational parameters to improve the quality of the approximation. We demonstrate experimentally that this approximation is much faster than sampling, but comparable in accuracy. We also introduce a simple new technique for handling evidence, which allows us to handle arbitrary distributionson observed nodes, as well as achieving a significant speedup in networks with discrete variables of large cardinality. 1 Introduction Many probabilistic models naturally contain discrete and continuous variables. (Such models are sometimes called "hybrid".) Unfortunately, exact inference is only possible when all the continuous variables are Gaussian and have no discrete children. If we want to al..