Rapid Mixing of Hamiltonian Monte Carlo on Strongly Log-Concave Distributions

Abstract

We obtain several quantitative bounds on the mixing properties of the Hamiltonian Monte Carlo (HMC) algorithm for a strongly log-concave target distribution $\pi$ on ${\mathbb{R}}^d$, showing that HMC mixes quickly in this setting. One of our main results is a dimension-free bound on the mixing of an "ideal" HMC chain, which is used to show that the usual leapfrog implementation of HMC can sample from $\pi$ using only $\mathcal{O}(d^{\frac{1}{4}})$ gradient evaluations. This dependence on dimension is sharp, and our results significantly extend and improve previous quantitative bounds on the mixing of HMC.

Cited by in corpus (15)

• Underdamped Langevin MCMC: A non-asymptotic analysis
• NeuTra-lizing Bad Geometry in Hamiltonian Monte Carlo Using Neural Transport
• Stacking for Non-mixing Bayesian Computations: The Curse and Blessing of Multimodal Posteriors
• Mixing Time Guarantees for Unadjusted Hamiltonian Monte Carlo
• Convergence of unadjusted Hamiltonian Monte Carlo for mean-field models
• Accelerating Langevin Sampling with Birth-death
• Optimal Convergence Rate of Hamiltonian Monte Carlo for Strongly Logconcave Distributions
• Proximal Langevin Algorithm: Rapid Convergence Under Isoperimetry
• Neural Bridge Sampling for Evaluating Safety-Critical Autonomous Systems
• Couplings for Andersen Dynamics
• An Introduction to Hamiltonian Monte Carlo Method for Sampling
• When is the Convergence Time of Langevin Algorithms Dimension Independent? A Composite Optimization Viewpoint
• Couplings for Multinomial Hamiltonian Monte Carlo
• Hamiltonian Monte Carlo Swindles
• Entropy-based adaptive Hamiltonian Monte Carlo