scispace - formally typeset
Open AccessJournal ArticleDOI

Reflection couplings and contraction rates for diffusions

TLDR
In this paper, the authors consider contractivity for diffusion semigroups w.r.t. Kantorovich distances based on appropriately chosen concave functions and show that with appropriate explicit choices of the underlying distance, contractivity with rates of close to optimal order can be obtained in several fundamental classes of examples.
Abstract
We consider contractivity for diffusion semigroups w.r.t. Kantorovich ($L^1$ Wasserstein) distances based on appropriately chosen concave functions. These distances are inbetween total variation and usual Wasserstein distances. It is shown that by appropriate explicit choices of the underlying distance, contractivity with rates of close to optimal order can be obtained in several fundamental classes of examples where contractivity w.r.t. standard Wasserstein distances fails. Applications include overdamped Langevin diffusions with locally non-convex potentials, products of these processes, and systems of weakly interacting diffusions, both of mean-field and nearest neighbour type.

read more

Citations
More filters
Journal ArticleDOI

Nonasymptotic convergence analysis for the unadjusted Langevin algorithm

TL;DR: In this article, a sampling technique based on the Euler discretization of the Langevin stochastic differential equation is studied, and for both constant and decreasing step sizes, non-asymptotic bounds for the convergence to stationarity in both total variation and Wasserstein distances are obtained.
Journal ArticleDOI

Efficient Bayesian computation by proximal Markov chain Monte Carlo: when Langevin meets Moreau

TL;DR: In this paper, a Markov chain Monte Carlo (MCMC) method is proposed for high-dimensional models that are log-concave and nonsmooth, a class of models that is central in imaging sciences.
Posted Content

Sharp Convergence Rates for Langevin Dynamics in the Nonconvex Setting.

TL;DR: Both overdamped and underdamped Langevin MCMC are studied and upper bounds on the number of steps required to obtain a sample from a distribution that is within $\epsilon$ of $p*$ in $1$-Wasserstein distance are established.

Supplement to "High-dimensional Bayesian inference via the Unadjusted Langevin Algorithm"

TL;DR: In this article, the authors proposed a method to solve the problem of the "missing link" problem in the context of biomedical data.http://hal.archives-ouvertes.fr/hal-01304430
Posted Content

Rapid Mixing of Hamiltonian Monte Carlo on Strongly Log-Concave Distributions

Oren Mangoubi, +1 more
- 23 Aug 2017 - 
TL;DR: In this paper, the mixing properties of the Hamiltonian Monte Carlo (HMC) algorithm for a strongly log-concave target distribution were studied and it was shown that HMC mixes quickly in this setting.
References
More filters
Book

Lectures on the Coupling Method

TL;DR: In this article, the authors propose Discrete Theory Continuous Theory, Inequalities Intensity-Governed Processes Diffusions Appendix Frequently Used Notation References Index, Section 5.
Book

Analysis and Geometry of Markov Diffusion Operators

TL;DR: Semigroups of bounded operators on a Banach space have been studied in this paper for Riemannian geometry and Markov semigroups have been used for stochastic calculus.
Journal ArticleDOI

Stochastic differential equations with reflecting boundary conditions

TL;DR: On resout des equations differentielles stochastiques a conditions aux limites reflechissantes par une approche directe basee sur le probleme de Skorokhod as discussed by the authors.
Book

Functional Integration And Partial Differential Equations

TL;DR: The authors discusses some aspects of the theory of partial differential equations from the viewpoint of probability theory and provides results that have not previously appeared in book form, including research contributions of the author.
Book

Stochastic analysis on manifolds

TL;DR: Brownian motion and analytic index theorems analysis on path spaces are studied in this article for stochastic differential equations and diffusions in the context of path spaces, where the heat kernel is used as an index.
Related Papers (5)