The Langevin Monte Carlo algorithm in the non-smooth log-concave case

DOI10.1214/23-AAP1935arXiv2101.10695MaRDI QIDQ6138925FDOQ6138925

Authors: Joseph Lehec 

Publication date: 16 January 2024

Published in: The Annals of Applied Probability (Search for Journal in Brave)

Abstract: We prove non asymptotic polynomial bounds on the convergence of the Langevin Monte Carlo algorithm in the case where the potential is a convex function which is globally Lipschitz on its domain, typically the maximum of a finite number of affine functions on an arbitrary convex set. In particular the potential is not assumed to be gradient Lipschitz, in contrast with most existing works on the topic.

Full work available at URL: https://arxiv.org/abs/2101.10695

zbMATH Keywords

Markov chain Monte Carloconvexitystatistical sampling

Mathematics Subject Classification ID

Monte Carlo methods (65C05) Sampling theory, sample surveys (62D05) Randomized algorithms (68W20) Asymptotic theory of convex bodies (52A23)

Cites Work

• Stochastic differential equations with reflecting boundary condition in convex regions
• Convex measures on locally convex spaces
• Title not available (Why is that?)
• Isoperimetric problems for convex bodies and a localization lemma
• A random polynomial-time algorithm for approximating the volume of convex bodies
• Generalization of an inequality by Talagrand and links with the logarithmic Sobolev inequality
• Hypercontractivity of Hamilton-Jacobi equations.
• Title not available (Why is that?)
• Analysis and geometry of Markov diffusion operators
• From Brunn-Minkowski to Brascamp-Lieb and to logarithmic Sobolev inequalities
• An almost constant lower bound of the isoperimetric coefficient in the KLS conjecture
• Title not available (Why is that?)
• Sections of convex bodies through their centroid
• From concentration to isoperimetry: semigroup proofs
• Isoperimetric and concentration inequalities: equivalence under curvature lower bound
• High-dimensional Bayesian inference via the unadjusted Langevin algorithm
• Thin shell implies spectral gap up to polylog via a stochastic localization scheme
• The Poincaré inequality and quadratic transportation-variance inequalities
• Theoretical Guarantees for Approximate Sampling from Smooth and Log-Concave Densities
• Sampling from a log-concave distribution with projected Langevin Monte Carlo
• Bourgain's slicing problem and KLS isoperimetry up to polylog
• A note on quadratic transportation and divergence inequality

Cited In (1)

• NF-ULA: normalizing flow-based unadjusted Langevin algorithm for imaging inverse problems

This page was built for publication: The Langevin Monte Carlo algorithm in the non-smooth log-concave case

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6138925)