Coupling and convergence for Hamiltonian Monte Carlo

DOI10.1214/19-AAP1528MaRDI QIDQ2657908

Nawaf Bou-Rabee, Andreas Eberle, Raphael Zimmer

Publication date: 18 March 2021

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

Full work available at URL: https://arxiv.org/abs/1805.00452

zbMATH Keywords

Markov chain Monte Carlogeometric integrationconvergence to equilibriumhybrid Monte CarlocouplingHamiltonian Monte Carlometropolis-Hastings

Mathematics Subject Classification ID

Computational methods in Markov chains (60J22) Monte Carlo methods (65C05) Numerical methods for Hamiltonian systems including symplectic integrators (65P10)

Related Items (37)

Finding the jump rate for fastest decay in the Goldstein-Taylor model ⋮ Improved bounds for discretization of Langevin diffusions: near-optimal rates without convexity ⋮ Quantitative contraction rates for Markov chains on general state spaces ⋮ Complexity of zigzag sampling algorithm for strongly log-concave distributions ⋮ Unnamed Item ⋮ Couplings for Andersen dynamics ⋮ Strict Kantorovich contractions for Markov chains and Euler schemes with general noise ⋮ On sampling from a log-concave density using kinetic Langevin diffusions ⋮ Coupling approach for exponential ergodicity of stochastic Hamiltonian systems with Lévy noises ⋮ On Irreversible Metropolis Sampling Related to Langevin Dynamics ⋮ Global Optimization via Schrödinger–Föllmer Diffusion ⋮ Geometric Ergodicity for Hamiltonian Monte Carlo on Compact Manifolds ⋮ Ergodicity and long-time behavior of the Random Batch Method for interacting particle systems ⋮ Unnamed Item ⋮ A blob method for inhomogeneous diffusion with applications to multi-agent control and sampling ⋮ Decentralized Bayesian learning with Metropolis-adjusted Hamiltonian Monte Carlo ⋮ Unbiased Estimation Using Underdamped Langevin Dynamics ⋮ Improving sampling accuracy of stochastic gradient MCMC methods via non-uniform subsampling of gradients ⋮ On Unbiased Estimation for Discretized Models ⋮ Convergence of unadjusted Hamiltonian Monte Carlo for mean-field models ⋮ The forward-backward envelope for sampling with the overdamped Langevin algorithm ⋮ Randomized time Riemannian manifold Hamiltonian Monte Carlo ⋮ Geometric numerical integration. Abstracts from the workshop held March 28 -- April 3, 2021 (hybrid meeting) ⋮ Simulating Coulomb and log-gases with hybrid Monte Carlo algorithms ⋮ High-dimensional MCMC with a standard splitting scheme for the underdamped Langevin diffusion ⋮ On the limitations of single-step drift and minorization in Markov chain convergence analysis ⋮ Mixing of Hamiltonian Monte Carlo on strongly log-concave distributions: continuous dynamics ⋮ Is there an analog of Nesterov acceleration for gradient-based MCMC? ⋮ Two-scale coupling for preconditioned Hamiltonian Monte Carlo in infinite dimensions ⋮ Constructing Sampling Schemes via Coupling: Markov Semigroups and Optimal Transport ⋮ Geometric integrators and the Hamiltonian Monte Carlo method ⋮ Markov chain simulation for multilevel Monte Carlo ⋮ Randomized Hamiltonian Monte Carlo as scaling limit of the bouncy particle sampler and dimension-free convergence rates ⋮ On \(L^2\) convergence of the Hamiltonian Monte Carlo ⋮ Approximations of piecewise deterministic Markov processes and their convergence properties ⋮ Mixing time guarantees for unadjusted Hamiltonian Monte Carlo ⋮ Wasserstein-based methods for convergence complexity analysis of MCMC with applications

Uses Software

• NUTS
• PRMLT

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo
• Reflection couplings and contraction rates for diffusions
• Equi-energy sampler with applications in statistical inference and statistical mechanics
• Curvature, concentration and error estimates for Markov chain Monte Carlo
• Asymptotic coupling and a general form of Harris' theorem with applications to stochastic delay equations
• Gradient estimates on manifolds using coupling
• An introduction to MCMC for machine learning
• Randomized Hamiltonian Monte Carlo
• Explicit contraction rates for a class of degenerate and infinite-dimensional diffusions
• Analysis of a nonreversible Markov chain sampler.
• Couplings and quantitative contraction rates for Langevin dynamics
• The geometric foundations of Hamiltonian Monte Carlo
• Error bounds for Metropolis-Hastings algorithms applied to perturbations of Gaussian measures in high dimensions
• Subgeometric rates of convergence of Markov processes in the Wasserstein metric
• On the rate of convergence to equilibrium of the Andersen thermostat in molecular dynamics
• Monte Carlo strategies in scientific computing.
• Analysis of SPDEs arising in path sampling. II: The nonlinear case
• The Markov chain Monte Carlo revolution
• Real-Parameter Evolutionary Monte Carlo With Applications to Bayesian Mixture Models
• Quantitative Harris-type theorems for diffusions and McKean–Vlasov processes
• Unbiased Hamiltonian Monte Carlo with couplings
• Geometric integrators and the Hamiltonian Monte Carlo method
• Markov Chain Monte Carlo and Numerical Differential Equations
• The Andersen thermostat in molecular dynamics

This page was built for publication: Coupling and convergence for Hamiltonian Monte Carlo