Control variates for stochastic gradient MCMC

Baker, Jack and Fearnhead, Paul and Fox, Emily B. and Nemeth, Christopher (2019) Control variates for stochastic gradient MCMC. Statistics and Computing, 29 (3). pp. 599-615. ISSN 0960-3174

[thumbnail of 1706.05439.pd]
Preview
PDF (1706.05439.pd)
1706.05439.pd.pdf - Accepted Version
Available under License Creative Commons Attribution.

Download (1MB)

Abstract

It is well known that Markov chain Monte Carlo (MCMC) methods scale poorly with dataset size. A popular class of methods for solving this issue is stochastic gradient MCMC (SGMCMC). These methods use a noisy estimate of the gradient of the log-posterior, which reduces the per iteration computational cost of the algorithm. Despite this, there are a number of results suggesting that stochastic gradient Langevin dynamics (SGLD), probably the most popular of these methods, still has computational cost proportional to the dataset size. We suggest an alternative log-posterior gradient estimate for stochastic gradient MCMC which uses control variates to reduce the variance. We analyse SGLD using this gradient estimate, and show that, under log-concavity assumptions on the target distribution, the computational cost required for a given level of accuracy is independent of the dataset size. Next we show that a different control variate technique, known as zero variance control variates, can be applied to SGMCMC algorithms for free. This post-processing step improves the inference of the algorithm by reducing the variance of the MCMC output. Zero variance control variates rely on the gradient of the log-posterior; we explore how the variance reduction is affected by replacing this with the noisy gradient estimate calculated by SGMCMC.

Item Type:
Journal Article
Journal or Publication Title:
Statistics and Computing
Additional Information:
The final publication is available at Springer via https://doi.org/10.1007/s11222-018-9826-2
Uncontrolled Keywords:
/dk/atira/pure/subjectarea/asjc/1700/1703
Subjects:
?? stat.cocs.lgstat.mlcomputational theory and mathematicstheoretical computer sciencestatistics and probabilitystatistics, probability and uncertainty ??
ID Code:
86844
Deposited By:
Deposited On:
26 Jun 2017 13:46
Refereed?:
Yes
Published?:
Published
Last Modified:
13 Aug 2024 23:36