Diffusion approximations and control variates for MCMC

Abstract : A new methodology is presented for the construction of control variates to reduce the variance of additive functionals of Markov Chain Monte Carlo (MCMC) samplers. Our control variates are defined as linear combinations of functions whose coefficients are obtained by minimizing a proxy for the asymptotic variance. The construction is theoretically justified by two new results. We first show that the asymptotic variances of some well-known MCMC algorithms, including the Random Walk Metropolis and the (Metropolis) Unadjusted/Adjusted Langevin Algorithm, are close to the asymptotic variance of the Langevin diffusion. Second, we provide an explicit representation of the optimal coefficients minimizing the asymptotic variance of the Langevin diffusion. Several examples of Bayesian inference problems demonstrate that the corresponding reduction in the variance is significant, and that in some cases it can be dramatic.
Complete list of metadatas

Cited literature [42 references]  Display  Hide  Download

https://hal.inria.fr/hal-01934316
Contributor : Nicolas Brosse <>
Submitted on : Sunday, November 25, 2018 - 10:01:23 PM
Last modification on : Thursday, October 17, 2019 - 12:36:10 PM
Long-term archiving on : Tuesday, February 26, 2019 - 12:48:09 PM

File

main.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01934316, version 1

Citation

Nicolas Brosse, Alain Durmus, Sean Meyn, Éric Moulines. Diffusion approximations and control variates for MCMC. 2018. ⟨hal-01934316⟩

Share

Metrics

Record views

137

Files downloads

69