Mixing Time Guarantees for Unadjusted Hamiltonian Monte Carlo

05/03/2021
by   Nawaf Bou-Rabee, et al.
0

We provide quantitative upper bounds on the total variation mixing time of the Markov chain corresponding to the unadjusted Hamiltonian Monte Carlo (uHMC) algorithm. For two general classes of models and fixed time discretization step size h, the mixing time is shown to depend only logarithmically on the dimension. Moreover, we provide quantitative upper bounds on the total variation distance between the invariant measure of the uHMC chain and the true target measure. As a consequence, we show that an ε-accurate approximation of the target distribution μ in total variation distance can be achieved by uHMC for a broad class of models with O(d^3/4ε^-1/2log (d/ε )) gradient evaluations, and for mean field models with weak interactions with O(d^1/2ε^-1/2log (d/ε )) gradient evaluations. The proofs are based on the construction of successful couplings for uHMC that realize the upper bounds.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
08/23/2017

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

We obtain several quantitative bounds on the mixing properties of the Ha...
research
08/22/2023

Nonlinear Hamiltonian Monte Carlo its Particle Approximation

We present a nonlinear (in the sense of McKean) generalization of Hamilt...
research
08/02/2021

Asymptotic bias of inexact Markov Chain Monte Carlo methods in high dimension

This paper establishes non-asymptotic bounds on Wasserstein distances be...
research
06/09/2019

The Implicit Metropolis-Hastings Algorithm

Recent works propose using the discriminator of a GAN to filter out unre...
research
06/15/2023

Second order quantitative bounds for unadjusted generalized Hamiltonian Monte Carlo

This paper provides a convergence analysis for generalized Hamiltonian M...
research
02/10/2020

Logsmooth Gradient Concentration and Tighter Runtimes for Metropolized Hamiltonian Monte Carlo

We show that the gradient norm ∇ f(x) for x ∼(-f(x)), where f is strongl...
research
04/10/2023

When does Metropolized Hamiltonian Monte Carlo provably outperform Metropolis-adjusted Langevin algorithm?

We analyze the mixing time of Metropolized Hamiltonian Monte Carlo (HMC)...

Please sign up or login with your details

Forgot password? Click here to reset