Resolving the Mixing Time of the Langevin Algorithm to its Stationary Distribution for Log-Concave Sampling

10/16/2022
by   Jason M. Altschuler, et al.
0

Sampling from a high-dimensional distribution is a fundamental task in statistics, engineering, and the sciences. A canonical approach is the Langevin Algorithm, i.e., the Markov chain for the discretized Langevin Diffusion. This is the sampling analog of Gradient Descent. Despite being studied for several decades in multiple communities, tight mixing bounds for this algorithm remain unresolved even in the seemingly simple setting of log-concave distributions over a bounded domain. This paper completely characterizes the mixing time of the Langevin Algorithm to its stationary distribution in this setting (and others). This mixing result can be combined with any bound on the discretization bias in order to sample from the stationary distribution of the continuous Langevin Diffusion. In this way, we disentangle the study of the mixing and bias of the Langevin Algorithm. Our key insight is to introduce a technique from the differential privacy literature to the sampling literature. This technique, called Privacy Amplification by Iteration, uses as a potential a variant of Rényi divergence that is made geometrically aware via Optimal Transport smoothing. This gives a short, simple proof of optimal mixing bounds and has several additional appealing properties. First, our approach removes all unnecessary assumptions required by other sampling analyses. Second, our approach unifies many settings: it extends unchanged if the Langevin Algorithm uses projections, stochastic mini-batch gradients, or strongly convex potentials (whereby our mixing time improves exponentially). Third, our approach exploits convexity only through the contractivity of a gradient step – reminiscent of how convexity is used in textbook proofs of Gradient Descent. In this way, we offer a new approach towards further unifying the analyses of optimization and sampling algorithms.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
01/08/2018

Log-concave sampling: Metropolis-Hastings algorithms are fast!

We consider the problem of sampling from a strongly log-concave density ...
research
12/24/2022

Concentration of the Langevin Algorithm's Stationary Distribution

A canonical algorithm for log-concave sampling is the Langevin Algorithm...
research
10/30/2020

Efficient constrained sampling via the mirror-Langevin algorithm

We propose a new discretization of the mirror-Langevin diffusion and giv...
research
05/27/2022

Privacy of Noisy Stochastic Gradient Descent: More Iterations without More Privacy Loss

A central issue in machine learning is how to train models on sensitive ...
research
11/29/2018

Simulated Tempering Langevin Monte Carlo II: An Improved Proof using Soft Markov Chain Decomposition

A key task in Bayesian machine learning is sampling from distributions t...
research
12/01/2021

On Mixing Times of Metropolized Algorithm With Optimization Step (MAO) : A New Framework

In this paper, we consider sampling from a class of distributions with t...
research
02/21/2019

Online Sampling from Log-Concave Distributions

Given a sequence of convex functions f_0, f_1, ..., f_T, we study the pr...

Please sign up or login with your details

Forgot password? Click here to reset