Hamiltonian-Assisted Metropolis Sampling

Zexi Song, Zhiqiang Tan

Research output: Contribution to journalArticlepeer-review


Various Markov chain Monte Carlo (MCMC) methods are studied to improve upon random walk Metropolis sampling, for simulation from complex distributions. Examples include Metropolis-adjusted Langevin algorithms, Hamiltonian Monte Carlo, and other algorithms related to underdamped Langevin dynamics. We propose a broad class of irreversible sampling algorithms, called Hamiltonian-assisted Metropolis sampling (HAMS), and develop two specific algorithms with appropriate tuning and preconditioning strategies. Our HAMS algorithms are designed to simultaneously achieve two distinctive properties, while using an augmented target density with a momentum as an auxiliary variable. One is generalized detailed balance, which induces an irreversible exploration of the target. The other is a rejection-free property for a Gaussian target with a prespecified variance matrix. This property allows our preconditioned algorithms to perform satisfactorily with relatively large step sizes. Furthermore, we formulate a framework of generalized Metropolis–Hastings sampling, which not only highlights our construction of HAMS at a more abstract level, but also facilitates possible further development of irreversible MCMC algorithms. We present several numerical experiments, where the proposed algorithms consistently yield superior results among existing algorithms using the same preconditioning schemes.

Original languageAmerican English
JournalJournal of the American Statistical Association
StateAccepted/In press - 2021

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty


  • Auxiliary variables
  • Detailed balance
  • Hamiltonian Monte Carlo
  • Markov chain Monte Carlo
  • Metropolis-adjusted Langevin algorithms
  • Metropolis–Hastings sampling
  • Underdamped Langevin dynamics


Dive into the research topics of 'Hamiltonian-Assisted Metropolis Sampling'. Together they form a unique fingerprint.

Cite this