Litcius/Paper detail

An entropic approach for Hamiltonian Monte Carlo: The idealized case

Pierre Monmarché

2024The Annals of Applied Probability12 citationsDOI

Abstract

Quantitative long-time entropic convergence and short-time regularization are established for an idealized Hamiltonian Monte Carlo chain which alternatively follows an Hamiltonian dynamics for a fixed time and then partially or totally refreshes its velocity with an auto-regressive Gaussian step. These results, in discrete time, are the analogues of similar results for the continuous-time kinetic Langevin diffusion, and the latter can be obtained from our bounds in a suitable limit regime. The dependency in the log-Sobolev constant of the target measure is sharp and is illustrated on a mean-field case and on a low-temperature regime, with an application to the simulated annealing algorithm. The practical unadjusted algorithm is briefly discussed.

Topics & Concepts

Statistical physicsMonte Carlo methodHamiltonian (control theory)GaussianHybrid Monte CarloMathematicsLangevin dynamicsPhysicsApplied mathematicsMarkov chain Monte CarloMathematical optimizationQuantum mechanicsStatisticsMarkov Chains and Monte Carlo MethodsStatistical Mechanics and EntropyAdvanced Thermodynamics and Statistical Mechanics
An entropic approach for Hamiltonian Monte Carlo: The idealized case | Litcius