Optimal scaling and diffusion limits for the Langevin algorithm in high dimensions

DOI10.1214/11-AAP828zbMath1272.60053arXiv1103.0542MaRDI QIDQ1931320

Natesh S. Pillai, Andrew M. Stuart, Alexandre H. Thiery

Publication date: 25 January 2013

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

Full work available at URL: https://arxiv.org/abs/1103.0542

zbMATH Keywords

diffusion approximation; Markov chain Monte Carlo; scaling limit; metropolis-adjusted Langevin algorithm

Mathematics Subject Classification ID

65C05: Monte Carlo methods

60J20: Applications of Markov chains and discrete-time Markov processes on general state spaces (social mobility, learning theory, industrial processes, etc.)

Related Items

Bayesian Inference Using Intermediate Distribution Based on Coarse Multiscale Model for Time Fractional Diffusion Equations, Some Remarks on Preconditioning Molecular Dynamics, Unnamed Item, A Bayesian Approach to Estimating Background Flows from a Passive Scalar, Asymptotic variance for random walk Metropolis chains in high dimensions: logarithmic growth via the Poisson equation, A function space HMC algorithm with second order Langevin diffusion limit, Bayesian computation: a summary of the current state, and samples backwards and forwards, MCMC methods for functions: modifying old algorithms to make them faster, Stochastic Gradient Markov Chain Monte Carlo, Efficient and generalizable tuning strategies for stochastic gradient MCMC, Optimal scaling for the transient phase of Metropolis Hastings algorithms: the longtime behavior, Noisy gradient flow from a random walk in Hilbert space, Asymptotic analysis of the random walk metropolis algorithm on ridged densities, Designing simple and efficient Markov chain Monte Carlo proposal kernels, Multimodal, high-dimensional, model-based, Bayesian inverse problems with applications in biomechanics, Study of new rare event simulation schemes and their application to extreme scenario generation, An adaptive multiple-try Metropolis algorithm, On sampling from a log-concave density using kinetic Langevin diffusions, Optimal scaling of the MALA algorithm with irreversible proposals for Gaussian targets, Optimal scaling of random-walk Metropolis algorithms on general target distributions, Non-asymptotic guarantees for sampling by stochastic gradient descent, Non-stationary phase of the MALA algorithm, Diffusion limit for the random walk Metropolis algorithm out of stationarity, On the efficiency of pseudo-marginal random walk Metropolis algorithms, Optimal scaling for the transient phase of the random walk Metropolis algorithm: the mean-field limit, Dimension-independent likelihood-informed MCMC, Scalable posterior approximations for large-scale Bayesian inverse problems via likelihood-informed parameter and state reduction, Optimal tuning of the hybrid Monte Carlo algorithm, Error bounds for Metropolis-Hastings algorithms applied to perturbations of Gaussian measures in high dimensions, On the stability of sequential Monte Carlo methods in high dimensions, Proximal Markov chain Monte Carlo algorithms, Randomize-Then-Optimize for Sampling and Uncertainty Quantification in Electrical Impedance Tomography, Accelerated Dimension-Independent Adaptive Metropolis, Sampling Constrained Probability Distributions Using Spherical Augmentation

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Diffusion limits of the random walk Metropolis algorithm in high dimensions
• Besov priors for Bayesian inverse problems
• Hybrid Monte Carlo on Hilbert spaces
• The random walk Metropolis: linking theory and practice through a case study
• Asymptotic behaviour of a class of stochastic approximation procedures
• Weak convergence and optimal scaling of random walk Metropolis algorithms
• Optimal scaling for various Metropolis-Hastings algorithms.
• From Metropolis to diffusions: Gibbs states and optimal scaling.
• Optimal scaling of MaLa for nonlinear regression.
• Discretization-invariant Bayesian inversion and Besov space priors
• Optimal scalings for local Metropolis-Hastings chains on nonproduct targets in high dimensions
• Optimal tuning of the hybrid Monte Carlo algorithm
• Weak convergence of Metropolis algorithms for non-I.I.D. target distributions
• Analysis of SPDEs arising in path sampling. II: The nonlinear case
• Analysis of SPDEs arising in path sampling. I: The Gaussian case
• Sparse deterministic approximation of Bayesian inverse problems
• Inverse problems: A Bayesian perspective
• MCMC METHODS FOR DIFFUSION BRIDGES
• Optimal Scaling of Discrete Approximations to Langevin Diffusions
• Scaling Limits for the Transient Phase of Local Metropolis–Hastings Algorithms
• Stochastic Equations in Infinite Dimensions