Optimal scaling for the transient phase of the random walk Metropolis algorithm: the mean-field limit

DOI10.1214/14-AAP1048MaRDI QIDQ2354897zbMATH OpenFDO

Authors Benjamin Jourdain, Błażej Miasojedow, Tony Lelièvre

Publication date 27 July 2015

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

Full work available at URL https://arxiv.org/abs/1210.7639, https://projecteuclid.org/euclid.aoap/1432212442

zbMATH Keywords

diffusion limitsmean-field limitoptimal scalingpropagation of chaosrandom walk Metropolis algorithm

Mathematics Subject Classification ID

Functional limit theorems; invariance principles (60F17) Exchangeability for stochastic processes (60G09) Sums of independent random variables; random walks (60G50) Markov chains (discrete-time Markov processes on discrete state spaces) (60J10) Computational methods in Markov chains (60J22) Diffusion processes (60J60) Monte Carlo methods (65C05) Numerical analysis or methods applied to Markov chains (65C40)

Abstract: We consider the random walk Metropolis algorithm on $mathbb{R}^n$ with Gaussian proposals, and when the target probability measure is the $n$-fold product of a one-dimensional law. In the limit $noinfty$, it is well known (see [Ann. Appl. Probab. 7 (1997) 110-120]) that, when the variance of the proposal scales inversely proportional to the dimension $n$ whereas time is accelerated by the factor $n$, a diffusive limit is obtained for each component of the Markov chain if this chain starts at equilibrium. This paper extends this result when the initial distribution is not the target probability measure. Remarking that the interaction between the components of the chain due to the common acceptance/rejection of the proposed moves is of mean-field type, we obtain a propagation of chaos result under the same scaling as in the stationary case. This proves that, in terms of the dimension $n$, the same scaling holds for the transient phase of the Metropolis-Hastings algorithm as near stationarity. The diffusive and mean-field limit of each component is a diffusion process nonlinear in the sense of McKean. This opens the route to new investigations of the optimal choice for the variance of the proposal distribution in order to accelerate convergence to equilibrium (see [Optimal scaling for the transient phase of Metropolis-Hastings algorithms: The longtime behavior Bernoulli (2014) To appear]).

Recommendations

• Optimal scaling for the transient phase of Metropolis Hastings algorithms: the longtime behavior
• Optimal scaling of the random walk Metropolis algorithm under L^p mean differentiability
• Optimal scaling of random-walk Metropolis algorithms on general target distributions
• Optimal scaling of random walk Metropolis algorithms with discontinuous target densities
• Scaling Limits for the Transient Phase of Local Metropolis–Hastings Algorithms
• Optimal scaling for the pseudo-marginal random walk Metropolis: insensitivity to the noise generating mechanism
• Optimal scaling of random walk Metropolis algorithms with non-Gaussian proposals
• Weak convergence and optimal scaling of random walk Metropolis algorithms
• Optimal scaling of the random walk Metropolis: general criteria for the 0.234 acceptance rule

Cites work

• Diffusion limits of the random walk Metropolis algorithm in high dimensions
• From Metropolis to diffusions: Gibbs states and optimal scaling.
• scientific article; zbMATH DE number 3951715 (Why is no real title available?)
• scientific article; zbMATH DE number 192908 (Why is no real title available?)
• Mass transportation problems. Vol. 1: Theory. Vol. 2: Applications
• Monte Carlo sampling methods using Markov chains and their applications
• Noisy gradient flow from a random walk in Hilbert space
• Optimal acceptance rates for Metropolis algorithms: Moving beyond 0.234
• Optimal scaling and diffusion limits for the Langevin algorithm in high dimensions
• Optimal scaling for the transient phase of Metropolis Hastings algorithms: the longtime behavior
• Optimal scaling for various Metropolis-Hastings algorithms.
• Optimal Scaling of Discrete Approximations to Langevin Diffusions
• Optimal scaling of MaLa for nonlinear regression.
• Optimal scaling of random walk Metropolis algorithms with discontinuous target densities
• Optimal scaling of random walk Metropolis algorithms with non-Gaussian proposals
• Optimal scalings for local Metropolis-Hastings chains on nonproduct targets in high dimensions
• Optimal tuning of the hybrid Monte Carlo algorithm
• Scaling analysis of delayed rejection MCMC methods
• Scaling analysis of multiple-try MCMC methods
• Scaling Limits for the Transient Phase of Local Metropolis–Hastings Algorithms
• Weak convergence and optimal scaling of random walk Metropolis algorithms
• Weak convergence of Metropolis algorithms for non-I.I.D. target distributions

Cited in

(19)

• From Metropolis to diffusions: Gibbs states and optimal scaling.
• A piecewise deterministic scaling limit of lifted Metropolis-Hastings in the Curie-Weiss model
• Propagation of chaos: a review of models, methods and applications. II: Applications
• Collective proposal distributions for nonlinear MCMC samplers: mean-field theory and fast implementation
• Optimal scaling of the MALA algorithm with irreversible proposals for Gaussian targets
• Optimal scaling of random-walk Metropolis algorithms on general target distributions
• Counterexamples for optimal scaling of Metropolis-Hastings chains with rough target densities
• Random walk Metropolis algorithm in high dimension with non-Gaussian target distributions
• Non-stationary phase of the MALA algorithm
• Diffusion limit for the random walk Metropolis algorithm out of stationarity
• Complexity bounds for Markov chain Monte Carlo algorithms via diffusion limits
• Optimal scaling of the random walk Metropolis algorithm under L^p mean differentiability
• Optimal scaling for the transient phase of Metropolis Hastings algorithms: the longtime behavior
• Derivative-Free Bayesian Inversion Using Multiscale Dynamics
• Asymptotic variance for random walk Metropolis chains in high dimensions: logarithmic growth via the Poisson equation
• Scaling Limits for the Transient Phase of Local Metropolis–Hastings Algorithms
• Bayesian computation: a summary of the current state, and samples backwards and forwards
• Singular relaxation of a random walk in a box with a Metropolis Monte Carlo dynamics
• Metropolis Monte Carlo sampling: convergence, localization transition and optimality

This page was built for publication: Optimal scaling for the transient phase of the random walk Metropolis algorithm: the mean-field limit

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2354897)