English

Optimal friction matrix for underdamped Langevin sampling

Computation 2023-11-01 v1 Probability

Abstract

A systematic procedure for optimising the friction coefficient in underdamped Langevin dynamics as a sampling tool is given by taking the gradient of the associated asymptotic variance with respect to friction. We give an expression for this gradient in terms of the solution to an appropriate Poisson equation and show that it can be approximated by short simulations of the associated first variation/tangent process under concavity assumptions on the log density. Our algorithm is applied to the estimation of posterior means in Bayesian inference problems and reduced variance is demonstrated when compared to the original underdamped and overdamped Langevin dynamics in both full and stochastic gradient cases.

Keywords

Cite

@article{arxiv.2112.06844,
  title  = {Optimal friction matrix for underdamped Langevin sampling},
  author = {Martin Chak and Nikolas Kantas and Tony Lelièvre and Grigorios A. Pavliotis},
  journal= {arXiv preprint arXiv:2112.06844},
  year   = {2023}
}

Comments

44 pages, 6 figures

R2 v1 2026-06-24T08:15:27.039Z