Coupling and convergence for hamiltonian monte carlo

NAWAF BOU-RABEE, ANDREAS EBERLE, RAPHAEL ZIMMER

Research output: Contribution to journalArticlepeer-review

3 Scopus citations

Abstract

Based on a new coupling approach, we prove that the transition step of the Hamiltonian Monte Carlo algorithm is contractive w.r.t. a carefully designed Kantorovich (L1 Wasserstein) distance. The lower bound for the contraction rate is explicit. Global convexity of the potential is not required, and thus multimodal target distributions are included. Explicit quantitative bounds for the number of steps required to approximate the stationary distribution up to a given error ϵ are a direct consequence of contractivity. These bounds show that HMC can overcome diffusive behavior if the duration of the Hamiltonian dynamics is adjusted appropriately.

Original languageEnglish (US)
Pages (from-to)1209-1250
Number of pages42
JournalAnnals of Applied Probability
Volume30
Issue number3
DOIs
StatePublished - Jun 2020

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Keywords

  • Convergence to equilibrium
  • Coupling
  • Geometric integration
  • Hamiltonian Monte Carlo
  • Hybrid Monte Carlo
  • Markov chain Monte Carlo
  • Metropolis-Hastings

Fingerprint

Dive into the research topics of 'Coupling and convergence for hamiltonian monte carlo'. Together they form a unique fingerprint.

Cite this