Stochastic Annealing for Variational Inference

TL;DR

This paper evaluates a stochastic annealing strategy for variational inference, showing it can improve optimization in some Bayesian models but not all, compared to deterministic methods.

Contribution

It introduces and empirically assesses a stochastic annealing approach for variational inference, comparing its effectiveness to deterministic annealing and no annealing.

Findings

Stochastic annealing improves results for GMM and HMM.

Deterministic annealing performs better on LDA.

Performance varies depending on the model.

Abstract

We empirically evaluate a stochastic annealing strategy for Bayesian posterior optimization with variational inference. Variational inference is a deterministic approach to approximate posterior inference in Bayesian models in which a typically non-convex objective function is locally optimized over the parameters of the approximating distribution. We investigate an annealing method for optimizing this objective with the aim of finding a better local optimal solution and compare with deterministic annealing methods and no annealing. We show that stochastic annealing can provide clear improvement on the GMM and HMM, while performance on LDA tends to favor deterministic annealing methods.