Convergence of unadjusted Hamiltonian Monte Carlo for mean-field models

Abstract

We present dimension-free convergence and discretization error bounds for the unadjusted Hamiltonian Monte Carlo algorithm applied to high-dimensional probability distributions of mean-field type. These bounds require the discretization step to be sufficiently small, but do not require strong convexity of either the unary or pairwise potential terms present in the mean-field model. To handle high dimensionality, our proof uses a particlewise coupling that is contractive in a complementary particlewise metric.

38 pages, 4 figures

References in corpus (6)

• Optimal scalings for local Metropolis--Hastings chains on nonproduct targets in high dimensions
• Rapid Mixing of Hamiltonian Monte Carlo on Strongly Log-Concave Distributions
• Dimension-free path-integral molecular dynamics without preconditioning
• Uniform long-time and propagation of chaos estimates for mean field kinetic particles in non-convex landscapes
• Two-scale coupling for preconditioned Hamiltonian Monte Carlo in infinite dimensions
• The Kinetic Fokker-planck Equation With Mean Field Interaction

Cited by in corpus (8)

• Mixing Time Guarantees for Unadjusted Hamiltonian Monte Carlo
• Couplings for Andersen Dynamics
• Mixing of Metropolis-Adjusted Markov Chains via Couplings: The High Acceptance Regime
• Particle Dual Averaging: Optimization of Mean Field Neural Networks with Global Convergence Rate Analysis
• Unadjusted Hamiltonian MCMC with Stratified Monte Carlo Time Integration
• Improving sampling efficacy on high dimensional distributions with thin high density regions using Conservative Hamiltonian Monte Carlo
• An entropic approach for Hamiltonian Monte Carlo: the idealized case
• Randomized Runge-Kutta-Nyström Methods for Unadjusted Hamiltonian and Kinetic Langevin Monte Carlo