Mixing time guarantees for unadjusted Hamiltonian Monte Carlo

DOI10.3150/21-BEJ1450MaRDI QIDQ2108472

Nawaf Bou-Rabee, Andreas Eberle

Publication date: 19 December 2022

Published in: Bernoulli (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2105.00887

zbMATH Keywords

mixing timecouplingsMCMCvariational integratorsHamiltonian Monte Carlo

Mathematics Subject Classification ID

Statistics (62-XX) Probability theory and stochastic processes (60-XX)

Related Items (12)

Particle dual averaging: optimization of mean field neural network with global convergence rate analysis* ⋮ Connecting the Dots: Numerical Randomized Hamiltonian Monte Carlo with State-Dependent Event Rates ⋮ Geometric numerical integration. Abstracts from the workshop held March 28 -- April 3, 2021 (hybrid meeting) ⋮ An entropic approach for Hamiltonian Monte Carlo: the idealized case ⋮ On the dissipation of ideal Hamiltonian Monte Carlo sampler ⋮ Contraction and convergence rates for discretized kinetic Langevin dynamics ⋮ Mixing of Metropolis-adjusted Markov chains via couplings: the high acceptance regime ⋮ Asymptotic bias of inexact Markov chain Monte Carlo methods in high dimension ⋮ Tuning diagonal scale matrices for HMC ⋮ \(L^2\)-Wasserstein contraction for Euler schemes of elliptic diffusions and interacting particle systems ⋮ Contraction rate estimates of stochastic gradient kinetic Langevin integrators ⋮ Mixing time guarantees for unadjusted Hamiltonian Monte Carlo

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Reflection couplings and contraction rates for diffusions
• Kac's program in kinetic theory
• Quantitative bounds for Markov chain convergence: Wasserstein and total variation distances
• Asymptotic coupling and a general form of Harris' theorem with applications to stochastic delay equations
• Coupling of multidimensional diffusions by reflection
• On the \(\Gamma\)-convergence of discrete dynamics and variational integrators
• One-shot coupling for certain stochastic recursive sequences.
• Estimates of the determinant of a perturbed identity matrix
• Two-scale coupling for preconditioned Hamiltonian Monte Carlo in infinite dimensions
• Mixing time guarantees for unadjusted Hamiltonian Monte Carlo
• On explicit \(L^2\)-convergence rate estimate for piecewise deterministic Markov processes in MCMC algorithms
• High-dimensional MCMC with a standard splitting scheme for the underdamped Langevin diffusion
• Mixing of Hamiltonian Monte Carlo on strongly log-concave distributions: continuous dynamics
• High-dimensional Bayesian inference via the unadjusted Langevin algorithm
• Couplings and quantitative contraction rates for Langevin dynamics
• Ergodicity of the 2D Navier-Stokes equations with degenerate stochastic forcing
• Quantitative contraction rates for Markov chains on general state spaces
• Coupling and convergence for Hamiltonian Monte Carlo
• Convergence of Numerical Time-Averaging and Stationary Measures via Poisson Equations
• Geometric Numerical Integration
• Simulating Hamiltonian Dynamics
• Discrete mechanics and variational integrators
• Mathematical Aspects of Mixing Times in Markov Chains
• Markov Chains and Stochastic Stability
• Hypocoercivity
• Nuclear norm of higher-order tensors
• Analysis and Geometry of Markov Diffusion Operators
• Unbiased Hamiltonian Monte Carlo with couplings
• Geodesic walks in polytopes
• Convergence rate of Riemannian Hamiltonian Monte Carlo and faster polytope volume computation
• Geometric integrators and the Hamiltonian Monte Carlo method
• Hypocoercivity for linear kinetic equations conserving mass
• Optimal Transport
• Convergence of unadjusted Hamiltonian Monte Carlo for mean-field models

This page was built for publication: Mixing time guarantees for unadjusted Hamiltonian Monte Carlo