Stochastic Runge-Kutta Accelerates Langevin Monte Carlo and Beyond
Abstract
Sampling with Markov chain Monte Carlo methods typically amounts to discretizing some continuous-time dynamics with numerical integration. In this paper, we establish the convergence rate of sampling algorithms obtained by discretizing smooth It\^o diffusions exhibiting fast $2$-Wasserstein contraction, based on local deviation properties of the integration scheme. In particular, we study a sampling algorithm constructed by discretizing the overdamped Langevin diffusion with the method of stochastic Runge-Kutta. For strongly convex potentials that are smooth up to a certain order, its iterates converge to the target distribution in $2$-Wasserstein distance in $\tilde{\mathcal{O}}(d\epsilon^{-2/3})$ iterations. This improves upon the best-known rate for strongly log-concave sampling based on the overdamped Langevin equation using only the gradient oracle without adjustment. Additionally, we extend our analysis of stochastic Runge-Kutta methods to uniformly dissipative diffusions with possibly non-convex potentials and show they achieve better rates compared to the Euler-Maruyama scheme on the dependence on tolerance $\epsilon$. Numerical studies show that these algorithms lead to better stability and lower asymptotic errors.
Cite
Text
Li et al. "Stochastic Runge-Kutta Accelerates Langevin Monte Carlo and Beyond." Neural Information Processing Systems, 2019.Markdown
[Li et al. "Stochastic Runge-Kutta Accelerates Langevin Monte Carlo and Beyond." Neural Information Processing Systems, 2019.](https://mlanthology.org/neurips/2019/li2019neurips-stochastic/)BibTeX
@inproceedings{li2019neurips-stochastic,
title = {{Stochastic Runge-Kutta Accelerates Langevin Monte Carlo and Beyond}},
author = {Li, Xuechen and Wu, Yi and Mackey, Lester and Erdogdu, Murat A},
booktitle = {Neural Information Processing Systems},
year = {2019},
pages = {7748-7760},
url = {https://mlanthology.org/neurips/2019/li2019neurips-stochastic/}
}