Improved bounds for discretization of Langevin diffusions: near-optimal rates without convexity

DOI10.3150/21-BEJ1343zbMath1493.65021arXiv1907.11331MaRDI QIDQ2137032

Nicolas Flammarion, Martin J. Wainwright, Wenlong Mou, Bartlett, Peter L.

Publication date: 16 May 2022

Published in: Bernoulli (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1907.11331

zbMATH Keywords

Markov chain Monte Carlonon-asymptotic boundKL divergenceEuler-Maruyama discretizationLangevin diffusion

Mathematics Subject Classification ID

Monte Carlo methods (65C05) Stochastic ordinary differential equations (aspects of stochastic analysis) (60H10) Diffusion processes (60J60) Computational methods for stochastic equations (aspects of stochastic analysis) (60H35) Numerical solutions to stochastic differential and integral equations (65C30)

Related Items (4)

Global Optimization via Schrödinger–Föllmer Diffusion ⋮ Unadjusted Langevin algorithm with multiplicative noise: total variation and Wasserstein bounds ⋮ (Non)-penalized multilevel methods for non-uniformly log-concave distributions ⋮ Distributed event-triggered unadjusted Langevin algorithm for Bayesian learning

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Reflection couplings and contraction rates for diffusions
• Stochastic stability of differential equations. With contributions by G. N. Milstein and M. B. Nevelson
• Weighted Csiszár-Kullback-Pinsker inequalities and applications to transportation inequalities
• A simple proof of the Poincaré inequality for a large class of probability measures
• On polynomial mixing bounds for stochastic differential equations
• Exponential convergence of Langevin distributions and their discrete approximations
• Metastability in reversible diffusion processes. I: Sharp asymptotics for capacities and exit times
• Logarithmic Sobolev inequalities and stochastic Ising models
• On sampling from a log-concave density using kinetic Langevin diffusions
• User-friendly guarantees for the Langevin Monte Carlo with inaccurate gradient
• Measuring sample quality with diffusions
• High-dimensional Bayesian inference via the unadjusted Langevin algorithm
• Nonasymptotic convergence analysis for the unadjusted Langevin algorithm
• Optimal transport bounds between the time-marginals of a multidimensional diffusion and its Euler scheme
• Ergodicity for SDEs and approximations: locally Lipschitz vector fields and degenerate noise.
• Quantitative contraction rates for Markov chains on general state spaces
• Multilevel Monte Carlo method for ergodic SDEs without contractivity
• Coupling and convergence for Hamiltonian Monte Carlo
• Nonasymptotic bounds for sampling algorithms without log-concavity
• An Algorithmic Introduction to Numerical Simulation of Stochastic Differential Equations
• Convergence of Numerical Time-Averaging and Stationary Measures via Poisson Equations
• Some inequalities satisfied by the quantities of information of Fisher and Shannon
• A First Course in the Numerical Analysis of Differential Equations
• Pathwise accuracy and ergodicity of metropolized integrators for SDEs
• Logarithmic Sobolev Inequalities
• Entropy production and the rate of convergence to equilibrium for the Fokker-Planck equation
• Nonasymptotic mixing of the MALA algorithm
• Stochastic Processes and Applications
• Sampling can be faster than optimization
• Statistical Inference for the Population Landscape via Moment-Adjusted Stochastic Gradients
• Theoretical Guarantees for Approximate Sampling from Smooth and Log-Concave Densities
• Expansion of the global error for numerical schemes solving stochastic differential equations

This page was built for publication: Improved bounds for discretization of Langevin diffusions: near-optimal rates without convexity