Randomized Hamiltonian Monte Carlo
From MaRDI portal
equilibrium mean squared displacementgeometric ergodicityHamiltonian Monte Carlohybrid Monte Carlointegrated autocorrelation timeLyapunov functionsMarkov chain Monte Carlorandomization
Dynamical systems and their relations with probability theory and stochastic processes (37A50) Applications of stochastic analysis (to PDEs, etc.) (60H30) Computational methods in Markov chains (60J22) Continuous-time Markov processes on general state spaces (60J25) Sampling theory, sample surveys (62D05)
Abstract: Tuning the durations of the Hamiltonian flow in Hamiltonian Monte Carlo (also called Hybrid Monte Carlo) (HMC) involves a tradeoff between computational cost and sampling quality, which is typically challenging to resolve in a satisfactory way. In this article we present and analyze a randomized HMC method (RHMC), in which these durations are i.i.d. exponential random variables whose mean is a free parameter. We focus on the small time step size limit, where the algorithm is rejection-free and the computational cost is proportional to the mean duration. In this limit, we prove that RHMC is geometrically ergodic under the same conditions that imply geometric ergodicity of the solution to underdamped Langevin equations. Moreover, in the context of a multi-dimensional Gaussian distribution, we prove that the sampling efficiency of RHMC, unlike that of constant duration HMC, behaves in a regular way. This regularity is also verified numerically in non-Gaussian target distributions. Finally we suggest variants of RHMC for which the time step size is not required to be small.
Recommendations
- MCMC using Hamiltonian dynamics
- Randomized Hamiltonian Monte Carlo as scaling limit of the bouncy particle sampler and dimension-free convergence rates
- On the geometric ergodicity of Hamiltonian Monte Carlo
- Optimal tuning of the hybrid Monte Carlo algorithm
- Geometric integrators and the Hamiltonian Monte Carlo method
Cited in
(68)- Variational Hamiltonian Monte Carlo via score matching
- Adaptive multi-stage integrators for optimal energy conservation in molecular simulations
- Simulating Coulomb and log-gases with hybrid Monte Carlo algorithms
- Well-posedness and long time behavior of singular Langevin stochastic differential equations
- Irreducibility and geometric ergodicity of Hamiltonian Monte Carlo
- Subgeometric hypocoercivity for piecewise-deterministic Markov process Monte Carlo methods
- Two-scale coupling for preconditioned Hamiltonian Monte Carlo in infinite dimensions
- Peskun-Tierney ordering for Markovian Monte Carlo: beyond the reversible scenario
- Recycling intermediate steps to improve Hamiltonian Monte Carlo
- Randomized Hamiltonian Monte Carlo as scaling limit of the bouncy particle sampler and dimension-free convergence rates
- Cores for piecewise-deterministic Markov processes used in Markov chain Monte Carlo
- Split Hamiltonian Monte Carlo revisited
- Mixing rates for Hamiltonian Monte Carlo algorithms in finite and infinite dimensions
- Approximations of piecewise deterministic Markov processes and their convergence properties
- HMC: reducing the number of rejections by not using leapfrog and some results on the acceptance rate
- On explicit \(L^2\)-convergence rate estimate for piecewise deterministic Markov processes in MCMC algorithms
- Hypocoercivity with Schur complements
- Complexity of zigzag sampling algorithm for strongly log-concave distributions
- Couplings for Andersen dynamics
- A note on Fisher information hypocoercive decay for the linear Boltzmann equation
- Mixing of Hamiltonian Monte Carlo on strongly log-concave distributions: continuous dynamics
- Hypocoercivity of piecewise deterministic Markov process-Monte Carlo
- Modified Hamiltonian Monte Carlo for Bayesian inference
- On the geometric ergodicity of Hamiltonian Monte Carlo
- Couplings and quantitative contraction rates for Langevin dynamics
- Coupling and convergence for Hamiltonian Monte Carlo
- Coupling approach for exponential ergodicity of stochastic Hamiltonian systems with Lévy noises
- Geometric numerical integration. Abstracts from the workshop held March 28 -- April 3, 2021 (hybrid meeting)
- A Hamiltonian Monte Carlo Method for Non-Smooth Energy Sampling
- SPECTRWM: Spectral Random Walk Method for the Numerical Solution of Stochastic Partial Differential Equations
- Langevin dynamics with general kinetic energies
- Kinetic energy choice in Hamiltonian/hybrid Monte Carlo
- The connections between Lyapunov functions for some optimization algorithms and differential equations
- Weighted \(L^2\)-contractivity of Langevin dynamics with singular potentials
- scientific article; zbMATH DE number 7626757 (Why is no real title available?)
- Connecting the Dots: Numerical Randomized Hamiltonian Monte Carlo with State-Dependent Event Rates
- Monte Carlo Simulation on the Stiefel Manifold via Polar Expansion
- Maximum Conditional Entropy Hamiltonian Monte Carlo Sampler
- Constructing sampling schemes via coupling: Markov semigroups and optimal transport
- Geometric integrators and the Hamiltonian Monte Carlo method
- Birth–death dynamics for sampling: global convergence, approximations and their asymptotics
- Gradient-Based Markov Chain Monte Carlo for Bayesian Inference With Non-differentiable Priors
- Exponential ergodicity for damping Hamiltonian dynamics with state-dependent and non-local collisions
- Convergence of unadjusted Hamiltonian Monte Carlo for mean-field models
- The Apogee to Apogee Path Sampler
- Randomized time Riemannian manifold Hamiltonian Monte Carlo
- On the dissipation of ideal Hamiltonian Monte Carlo sampler
- An entropic approach for Hamiltonian Monte Carlo: the idealized case
- Accelerating MCMC algorithms
- Log-density gradient covariance and automatic metric tensors for Riemann manifold Monte Carlo methods
- Tuning diagonal scale matrices for HMC
- Stochastic automatic differentiation for Monte Carlo processes
- Non-reversible lifts of reversible diffusion processes and relaxation times
- Hypocoercivity meets lifts
- Hamiltonian Monte Carlo with asymmetrical momentum distributions
- Quantifying the effectiveness of linear preconditioning in Markov chain Monte Carlo
- Debiasing piecewise deterministic Markov process samplers using couplings
- Gradient-adjusted underdamped Langevin dynamics for sampling
- Numerical generalized randomized HMC processes for restricted domains
- Unadjusted Hamiltonian MCMC with stratified Monte Carlo time integration
- Hybrid Monte Carlo on Hilbert spaces
- On the convergence of dynamic implementations of Hamiltonian Monte Carlo and no U-turn samplers
- GIST: Gibbs self-tuning for locally adaptive Hamiltonian Monte Carlo
- Space-time divergence lemmas and optimal non-reversible lifts of diffusions on Riemannian manifolds with boundary
- Unbiased kinetic Langevin Monte Carlo with inexact gradients
- Adaptive tuning of Hamiltonian Monte Carlo methods
- Concentration inequalities and UQ bounds for hypocoercive MCMC samplers
- Faster Hamiltonian Monte Carlo by learning leapfrog scale: an offline randomized solution
This page was built for publication: Randomized Hamiltonian Monte Carlo
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1676436)
