Mini-batch Metropolis-Hastings MCMC with Reversible SGLD Proposal

Abstract

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.

References in corpus (12)

• Adam: A Method for Stochastic Optimization
• The pseudo-marginal approach for efficient Monte Carlo computations
• Stochastic Gradient Hamiltonian Monte Carlo
• Austerity in MCMC Land: Cutting the Metropolis-Hastings Budget
• Speeding Up MCMC by Efficient Data Subsampling
• Preconditioned Stochastic Gradient Langevin Dynamics for Deep Neural Networks
• Non-convex learning via Stochastic Gradient Langevin Dynamics: a nonasymptotic analysis
• On Markov chain Monte Carlo methods for tall data
• Parallelizing MCMC via Weierstrass Sampler
• A Walk with SGD
• Firefly Monte Carlo: Exact MCMC with Subsets of Data
• Mini-batch Tempered MCMC