Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 454 - Computational Advances in Approximate Bayesian Methods
Type: Topic Contributed
Date/Time: Thursday, August 6, 2020 : 10:00 AM to 11:50 AM
Sponsor: Section on Bayesian Statistical Science
Abstract #312210
Title: Mini-Batch Metropolis-Hastings MCMC with Reversible SGLD Proposal
Author(s): Rachel Wang* and Tung-Yu Wu and Wing Hung Wong
Companies: University of Sydney and Stanford University and Stanford University
Keywords: scalable MCMC; tempering; neural networks

Traditional MCMC algorithms are computationally intensive and do not scale well to large data. In particular, the Metropolis-Hastings (MH) algorithm requires passing over the entire dataset to evaluate the likelihood ratio in each iteration. We propose a general framework for performing MH-MCMC using mini-batches of the whole dataset and show that this gives rise to approximately a tempered stationary distribution. We prove that the algorithm preserves the modes of the original target distribution and derive an error bound on the approximation with mild assumptions on the likelihood. To further extend the utility of the algorithm to high dimensional settings, we construct a proposal with forward and reverse moves using stochastic gradient and show that the construction leads to reasonable acceptance probabilities. We demonstrate the performance of our algorithm in both low dimensional models and high dimensional neural network applications. Particularly in the latter case, compared to popular optimization methods, our method is more robust to the choice of learning rate and improves testing accuracy.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program