Variational approximations using Fisher divergence

Abstract

Modern applications of Bayesian inference involve models that are sufficiently complex that the corresponding posterior distributions are intractable and must be approximated. The most common approximation is based on Markov chain Monte Carlo, but these can be expensive when the data set is large and/or the model is complex, so more efficient variational approximations have recently received considerable attention. The traditional variational methods, that seek to minimize the Kullback--Leibler divergence between the posterior and a relatively simple parametric family, provide accurate and efficient estimation of the posterior mean, but often does not capture other moments, and have limitations in terms of the models to which they can be applied. Here we propose the construction of variational approximations based on minimizing the Fisher divergence, and develop an efficient computational algorithm that can be applied to a wide range of models without conjugacy or potentially unrealistic mean-field assumptions. We demonstrate the superior performance of the proposed method for the benchmark case of logistic regression.

13 pages, 5 figures, 2 tables

References in corpus (4)

• Rényi Divergence and Kullback-Leibler Divergence
• Variational inference for large-scale models of discrete choice
• Rényi Divergence Variational Inference
• Practical bounds on the error of Bayesian posterior approximations: A nonasymptotic approach

Cited by in corpus (4)

• Generalized Variational Inference: Three arguments for deriving new Posteriors
• Fast approximations of the Jeffreys divergence between univariate Gaussian mixture models via exponential polynomial densities
• Bi-level Score Matching for Learning Energy-based Latent Variable Models
• Multiple Instance Learning with Auxiliary Task Weighting for Multiple Myeloma Classification