A Contrastive Divergence for Combining Variational Inference and MCMC

Abstract

We develop a method to combine Markov chain Monte Carlo (MCMC) and variational inference (VI), leveraging the advantages of both inference approaches. Specifically, we improve the variational distribution by running a few MCMC steps. To make inference tractable, we introduce the variational contrastive divergence (VCD), a new divergence that replaces the standard Kullback-Leibler (KL) divergence used in VI. The VCD captures a notion of discrepancy between the initial variational distribution and its improved version (obtained after running the MCMC steps), and it converges asymptotically to the symmetrized KL divergence between the variational distribution and the posterior of interest. The VCD objective can be optimized efficiently with respect to the variational parameters via stochastic optimization. We show experimentally that optimizing the VCD leads to better predictive performance on two latent variable models: logistic matrix factorization and variational autoencoders (VAEs).

International Conference on Machine Learning (ICML 2019). 12 pages, 3 figures

Cited by in corpus (11)

• Data Augmentation in High Dimensional Low Sample Size Setting Using a Geometry-Based Variational Autoencoder
• A bi-partite generative model framework for analyzing and simulating large scale multiple discrete-continuous travel behaviour data
• Fast and Accurate Estimation of Non-Nested Binomial Hierarchical Models Using Variational Inference
• Markovian Score Climbing: Variational Inference with KL(p||q)
• Continual Multi-task Gaussian Processes
• All in the Exponential Family: Bregman Duality in Thermodynamic Variational Inference
• MCMC Variational Inference via Uncorrected Hamiltonian Annealing
• Variationally Inferred Sampling Through a Refined Bound for Probabilistic Programs
• Doubly Adaptive Importance Sampling
• MCMC-Interactive Variational Inference
• Contributions to Large Scale Bayesian Inference and Adversarial Machine Learning