Speeding Up MCMC by Efficient Data Subsampling

Abstract

We propose Subsampling MCMC, a Markov Chain Monte Carlo (MCMC) framework where the likelihood function for $n$ observations is estimated from a random subset of $m$ observations. We introduce a highly efficient unbiased estimator of the log-likelihood based on control variates, such that the computing cost is much smaller than that of the full log-likelihood in standard MCMC. The likelihood estimate is bias-corrected and used in two dependent pseudo-marginal algorithms to sample from a perturbed posterior, for which we derive the asymptotic error with respect to $n$ and $m$, respectively. We propose a practical estimator of the error and show that the error is negligible even for a very small $m$ in our applications. We demonstrate that Subsampling MCMC is substantially more efficient than standard MCMC in terms of sampling efficiency for a given computational budget, and that it outperforms other subsampling methods for MCMC proposed in the literature.

Main changes: The theory has been significantly revised

References in corpus (11)

• Searching for Exotic Particles in High-Energy Physics with Deep Learning
• The pseudo-marginal approach for efficient Monte Carlo computations
• On Markov chain Monte Carlo methods for tall data
• Parallelizing MCMC via Weierstrass Sampler
• Firefly Monte Carlo: Exact MCMC with Subsets of Data
• Speeding Up MCMC by Delayed Acceptance and Data Subsampling
• Coupled MCMC with a randomized acceptance probability
• Accelerating pseudo-marginal Metropolis-Hastings by correlating auxiliary variables
• The Block Pseudo-Marginal Sampler
• Hamiltonian Monte Carlo with Energy Conserving Subsampling
• The block-Poisson estimator for optimally tuned exact subsampling MCMC

Cited by in corpus (40)

• On Markov chain Monte Carlo methods for tall data
• Speeding Up MCMC by Delayed Acceptance and Data Subsampling
• Random projections for Bayesian regression
• Ergodicity of Approximate MCMC Chains with Applications to Large Data Sets
• Stacking for Non-mixing Bayesian Computations: The Curse and Blessing of Multimodal Posteriors
• A Survey of Bayesian Statistical Approaches for Big Data
• Approximating Bayes in the 21st Century
• Bayesian computation: a perspective on the current state, and sampling backwards and forwards
• Bayesian Conditional Density Filtering
• Hamiltonian Monte Carlo with Energy Conserving Subsampling
• The reproducing Stein kernel approach for post-hoc corrected sampling
• Computing Bayes: From Then 'Til Now'
• Optimal Distributed Subsampling for Maximum Quasi-Likelihood Estimators with Massive Data
• Fast Inference for Intractable Likelihood Problems using Variational Bayes
• Efficient MCMC Sampling with Dimension-Free Convergence Rate using ADMM-type Splitting
• Embarrassingly parallel MCMC using deep invertible transformations
• A survey of Monte Carlo methods for noisy and costly densities with application to reinforcement learning and ABC
• Distributed Computation for Marginal Likelihood based Model Choice
• Spectral Subsampling MCMC for Stationary Time Series
• Flexible Bayesian Nonlinear Model Configuration
• The block-Poisson estimator for optimally tuned exact subsampling MCMC
• Deep Bayesian regression models
• Asymptotically Optimal Exact Minibatch Metropolis-Hastings
• Divide-and-Conquer Bayesian Inference in Hidden Markov Models
• Accelerating Convergence of Replica Exchange Stochastic Gradient MCMC via Variance Reduction
• An efficient adaptive MCMC algorithm for Pseudo-Bayesian quantum tomography
• Mini-batch Metropolis-Hastings MCMC with Reversible SGLD Proposal
• Combining Model and Parameter Uncertainty in Bayesian Neural Networks
• Scalable Metropolis-Hastings for Exact Bayesian Inference with Large Datasets
• Incorporating Subsampling into Bayesian Models for High-Dimensional Spatial Data
• Most Likely Optimal Subsampled Markov Chain Monte Carlo
• Approximate Cross-validated Mean Estimates for Bayesian Hierarchical Regression Models
• Targeted stochastic gradient Markov chain Monte Carlo for hidden Markov models with rare latent states
• Variational Inference for Shrinkage Priors: The R package vir
• A Bayesian Hidden Semi-Markov Model with Covariate-Dependent State Duration Parameters for High-Frequency Data from Wearable Devices
• An Algorithm for Distributed Bayesian Inference in Generalized Linear Models
• What is the best predictor that you can compute in five minutes using a given Bayesian hierarchical model?
• Perturbation Analysis of Markov Chain Monte Carlo for Graphical Models
• Probabilistic Programs with Stochastic Conditioning
• On MCMC for variationally sparse Gaussian processes: A pseudo-marginal approach