Conditionally conjugate mean-field variational Bayes for logistic models

Abstract

Variational Bayes (VB) is a common strategy for approximate Bayesian inference, but simple methods are only available for specific classes of models including, in particular, representations having conditionally conjugate constructions within an exponential family. Models with logit components are an apparently notable exception to this class, due to the absence of conjugacy between the logistic likelihood and the Gaussian priors for the coefficients in the linear predictor. To facilitate approximate inference within this widely used class of models, Jaakkola and Jordan (2000) proposed a simple variational approach which relies on a family of tangent quadratic lower bounds of logistic log-likelihoods, thus restoring conjugacy between these approximate bounds and the Gaussian priors. This strategy is still implemented successfully, but less attempts have been made to formally understand the reasons underlying its excellent performance. To cover this key gap, we provide a formal connection between the above bound and a recent Pólya-gamma data augmentation for logistic regression. Such a result places the computational methods associated with the aforementioned bounds within the framework of variational inference for conditionally conjugate exponential family models, thereby allowing recent advances for this class to be inherited also by the methods relying on Jaakkola and Jordan (2000).

References in corpus (3)

• Variational Inference: A Review for Statisticians
• Sparse logistic principal components analysis for binary data
• Convergence and asymptotic normality of variational Bayesian approximations for exponential family models with missing values

Cited by in corpus (8)

• Stratified stochastic variational inference for high-dimensional network factor model
• Fast Variational Inference for Bayesian Factor Analysis in Single and Multi-Study Settings
• Hyperbolic Secant representation of the logistic function: Application to probabilistic Multiple Instance Learning for CT intracranial hemorrhage detection
• Partially factorized variational inference for high-dimensional mixed models
• Variational inference based on a subclass of closed skew normals
• A nested expectation-maximization algorithm for latent class models with covariates
• Variational Bayes latent class approach for EHR-based phenotyping with large real-world data
• Non-conjugate variational Bayes for pseudo-likelihood mixed effect models