Langevin diffusions and the Metropolis-adjusted Langevin algorithm

DOI10.1016/J.SPL.2014.04.002MaRDI QIDQ2453987zbMATH OpenOpenAlexWikidataFDO

Authors Yanyan Li

Publication date 12 June 2014

Published in Statistics \& Probability Letters (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1309.2983

zbMATH Keywords

Markov chain Monte CarloRiemannian manifoldsdiffusionsMetropolis-adjusted Langevin algorithm

Mathematics Subject Classification ID

Diffusion processes (60J60) Continuous-time Markov processes on general state spaces (60J25)

Abstract: We provide a clarification of the description of Langevin diffusions on Riemannian manifolds and of the measure underlying the invariant density. As a result we propose a new position-dependent Metropolis-adjusted Langevin algorithm (MALA) based upon a Langevin diffusion in $mathbb{R}^d$ which has the required invariant density with respect to Lebesgue measure. We show that our diffusion and the diffusion upon which a previously-proposed position-dependent MALA is based are equivalent in some cases but are distinct in general. A simulation study illustrates the gain in efficiency provided by the new position-dependent MALA.

Recommendations

• Smoothed Langevin proposals in Metropolis-Hastings algorithms.
• Langevin-type models. I: Diffusions with given stationary distributions and their discretizations
• Optimal Scaling of Discrete Approximations to Langevin Diffusions
• Higher order Langevin Monte Carlo algorithm
• An adaptive version for the Metropolis adjusted Langevin algorithm with a truncated drift

Cites work

• scientific article; zbMATH DE number 840151 (Why is no real title available?)
• scientific article; zbMATH DE number 3280855 (Why is no real title available?)
• Exponential convergence of Langevin distributions and their discrete approximations
• Higher-order implicit strong numerical schemes for stochastic differential equations
• Langevin diffusions and Metropolis-Hastings algorithms
• Optimal Scaling of Discrete Approximations to Langevin Diffusions
• Parameter Estimation for Differential Equations: a Generalized Smoothing Approach
• Riemann manifold Langevin and Hamiltonian Monte Carlo methods. With discussion and authors' reply
• Stochastic Calculus
• Stochastic differential equations. An introduction with applications.
• Time-reversible diffusions

Cited in

(44)

• Stochastic seismic waveform inversion using generative adversarial networks as a geological prior
• Langevin diffusions and Metropolis-Hastings algorithms
• The Langevin Equation as a Global Minimization Algorithm
• Uncertainty quantification in scientific machine learning: methods, metrics, and comparisons
• Stochastic gradient MCMC for state space models
• Adaptive step size selection for Hessian-based manifold Langevin samplers
• Stability of partially implicit Langevin schemes and their MCMC variants
• Bayesian inference for nonstationary marginal extremes
• Optimal volume-corrected Laplace-Metropolis method
• Computational efficiency study of a micro-macro Markov chain Monte Carlo method for molecular dynamics
• NeuralUQ: A Comprehensive Library for Uncertainty Quantification in Neural Differential Equations and Operators
• Measuring sample quality with diffusions
• Gradient-based adaptive importance samplers
• An adaptive Hessian approximated stochastic gradient MCMC method
• Information-geometric Markov chain Monte Carlo methods using diffusions
• Simulator-free solution of high-dimensional stochastic elliptic partial differential equations using deep neural networks
• Oracle lower bounds for stochastic gradient sampling algorithms
• Metropolis integration schemes for self-adjoint diffusions
• Bayesian computation: a summary of the current state, and samples backwards and forwards
• Multi-index antithetic stochastic gradient algorithm
• Exact targeting of Gibbs distributions using velocity-jump processes
• Autoregressive Moving Average Infinite Hidden Markov-Switching Models
• Geometric Integration of Measure-Preserving Flows for Sampling
• Hierarchical models: local proposal variances for RWM-within-Gibbs and MALA-within-Gibbs
• Some Remarks on Preconditioning Molecular Dynamics
• Lévy Langevin Monte Carlo
• Flexible covariate representations for extremes
• Irreversible samplers from jump and continuous Markov processes
• On Irreversible Metropolis Sampling Related to Langevin Dynamics
• Nonasymptotic bounds for sampling algorithms without log-concavity
• Particle Metropolis-adjusted Langevin algorithms
• Geometric MCMC for infinite-dimensional inverse problems
• Convergence of Position-Dependent MALA with Application to Conditional Simulation in GLMMs
• The forward-backward envelope for sampling with the overdamped Langevin algorithm
• Sampling from manifold-restricted distributions using tangent bundle projections
• Langevin diffusion for population based sampling with an application in Bayesian inference for pharmacodynamics
• Geometric adaptive Monte Carlo in random environment
• Adaptive dimension reduction to accelerate infinite-dimensional geometric Markov chain Monte Carlo
• On sampling from a log-concave density using kinetic Langevin diffusions
• Geometry-informed irreversible perturbations for accelerated convergence of Langevin dynamics
• Hierarchical models and tuning of random walk Metropolis algorithms
• Modified Cholesky Riemann manifold Hamiltonian Monte Carlo: exploiting sparsity for fast sampling of high-dimensional targets
• Smoothed Langevin proposals in Metropolis-Hastings algorithms.
• Langevin-type models. I: Diffusions with given stationary distributions and their discretizations

This page was built for publication: Langevin diffusions and the Metropolis-adjusted Langevin algorithm

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2453987)