Sharp convergence rates for Langevin dynamics in the nonconvex setting
Abstract
We study the problem of sampling from a distribution , where the function is -smooth everywhere and -strongly convex outside a ball of radius , but potentially nonconvex inside this ball. We study both overdamped and underdamped Langevin MCMC and establish upper bounds on the number of steps required to obtain a sample from a distribution that is within of in -Wasserstein distance. For the first-order method (overdamped Langevin MCMC), the iteration complexity is , where is the dimension of the underlying space. For the second-order method (underdamped Langevin MCMC), the iteration complexity is for an explicit positive constant . Surprisingly, the iteration complexity for both these algorithms is only polynomial in the dimension and the target accuracy . It is exponential, however, in the problem parameter , which is a measure of non-log-concavity of the target distribution.
Keywords
Cite
@article{arxiv.1805.01648,
title = {Sharp convergence rates for Langevin dynamics in the nonconvex setting},
author = {Xiang Cheng and Niladri S. Chatterji and Yasin Abbasi-Yadkori and Peter L. Bartlett and Michael I. Jordan},
journal= {arXiv preprint arXiv:1805.01648},
year = {2020}
}
Comments
78 pages, 2 figures