Variational Bayesian Inference with Stochastic Search

Abstract

Mean-field variational inference is a method for approximate Bayesian posterior inference. It approximates a full posterior distribution with a factorized set of distributions by maximizing a lower bound on the marginal likelihood. This requires the ability to integrate a sum of terms in the log joint likelihood using this factorized distribution. Often not all integrals are in closed form, which is typically handled by using a lower bound. We present an alternative algorithm based on stochastic optimization that allows for direct optimization of the variational lower bound. This method uses control variates to reduce the variance of the stochastic search gradient, in which existing lower bounds can play an important role. We demonstrate the approach on two non-conjugate models: logistic regression and an approximation to the HDP.

Appears in Proceedings of the 29th International Conference on Machine Learning (ICML 2012)

Cited by in corpus (11)

• Markov Chain Monte Carlo and Variational Inference: Bridging the Gap
• Deep and Confident Prediction for Time Series at Uber
• Q-Prop: Sample-Efficient Policy Gradient with An Off-Policy Critic
• Deep Probabilistic Programming
• Structured Stochastic Variational Inference
• Statistical Decision Making for Optimal Budget Allocation in Crowd Labeling
• Stochastic Variational Inference
• Latent Gaussian Processes for Distribution Estimation of Multivariate Categorical Data
• Local Expectation Gradients for Doubly Stochastic Variational Inference
• Online Inference for Relation Extraction with a Reduced Feature Set
• A Divergence Bound for Hybrids of MCMC and Variational Inference and an Application to Langevin Dynamics and SGVI