Unadjusted Hamiltonian MCMC with Stratified Monte Carlo Time Integration

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Bou-Rabee, Nawaf, Marsden, Milo
Format: Preprint
Veröffentlicht: 2022
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916634110722048
author Bou-Rabee, Nawaf
Marsden, Milo
author_facet Bou-Rabee, Nawaf
Marsden, Milo
contents A randomized time integrator is suggested for unadjusted Hamiltonian Monte Carlo (uHMC) which involves a very minor modification to the usual Verlet time integrator, and hence, is easy to implement. For target distributions of the form $μ(dx) \propto e^{-U(x)} dx$ where $U: \mathbb{R}^d \to \mathbb{R}_{\ge 0}$ is $K$-strongly convex but only $L$-gradient Lipschitz, and initial distributions $ν$ with finite second moment, coupling proofs reveal that an $\varepsilon$-accurate approximation of the target distribution in $L^2$-Wasserstein distance $\boldsymbol{\mathcal{W}}^2$ can be achieved by the uHMC algorithm with randomized time integration using $O\left((d/K)^{1/3} (L/K)^{5/3} \varepsilon^{-2/3} \log( \boldsymbol{\mathcal{W}}^2(μ, ν) / \varepsilon)^+\right)$ gradient evaluations; whereas for such rough target densities the corresponding complexity of the uHMC algorithm with Verlet time integration is in general $O\left((d/K)^{1/2} (L/K)^2 \varepsilon^{-1} \log( \boldsymbol{\mathcal{W}}^2(μ, ν) / \varepsilon)^+ \right)$. Metropolis-adjustable randomized time integrators are also provided.
format Preprint
id arxiv_https___arxiv_org_abs_2211_11003
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Unadjusted Hamiltonian MCMC with Stratified Monte Carlo Time Integration
Bou-Rabee, Nawaf
Marsden, Milo
Probability
Numerical Analysis
Statistics Theory
Computation
Machine Learning
60J05 (Primary) 65C05, 65P10 (Secondary)
A randomized time integrator is suggested for unadjusted Hamiltonian Monte Carlo (uHMC) which involves a very minor modification to the usual Verlet time integrator, and hence, is easy to implement. For target distributions of the form $μ(dx) \propto e^{-U(x)} dx$ where $U: \mathbb{R}^d \to \mathbb{R}_{\ge 0}$ is $K$-strongly convex but only $L$-gradient Lipschitz, and initial distributions $ν$ with finite second moment, coupling proofs reveal that an $\varepsilon$-accurate approximation of the target distribution in $L^2$-Wasserstein distance $\boldsymbol{\mathcal{W}}^2$ can be achieved by the uHMC algorithm with randomized time integration using $O\left((d/K)^{1/3} (L/K)^{5/3} \varepsilon^{-2/3} \log( \boldsymbol{\mathcal{W}}^2(μ, ν) / \varepsilon)^+\right)$ gradient evaluations; whereas for such rough target densities the corresponding complexity of the uHMC algorithm with Verlet time integration is in general $O\left((d/K)^{1/2} (L/K)^2 \varepsilon^{-1} \log( \boldsymbol{\mathcal{W}}^2(μ, ν) / \varepsilon)^+ \right)$. Metropolis-adjustable randomized time integrators are also provided.
title Unadjusted Hamiltonian MCMC with Stratified Monte Carlo Time Integration
topic Probability
Numerical Analysis
Statistics Theory
Computation
Machine Learning
60J05 (Primary) 65C05, 65P10 (Secondary)
url https://arxiv.org/abs/2211.11003