Hypocoercivity properties of adaptive Langevin dynamics
From MaRDI portal
Asymptotic behavior of solutions to PDEs (35B40) Fokker-Planck equations (35Q84) Applications of functional analysis in probability theory and statistics (46N30) Applications of Brownian motions and diffusion theory (population genetics, absorption problems, etc.) (60J70) Numerical solutions to stochastic differential and integral equations (65C30)
Abstract: Adaptive Langevin dynamics is a method for sampling the Boltzmann-Gibbs distribution at prescribed temperature in cases where the potential gradient is subject to stochastic perturbation of unknown magnitude. The method replaces the friction in underdamped Langevin dynamics with a dynamical variable, updated according to a negative feedback loop control law as in the Nos'e-Hoover thermostat. Using a hypocoercivity analysis we show that the law of Adaptive Langevin dynamics converges exponentially rapidly to the stationary distribution, with a rate that can be quantified in terms of the key parameters of the dynamics. This allows us in particular to obtain a central limit theorem with respect to the time averages computed along a stochastic path. Our theoretical findings are illustrated by numerical simulations involving classification of the MNIST data set of handwritten digits using Bayesian logistic regression.
Recommendations
Cites work
- A simple proof of the Poincaré inequality for a large class of probability measures
- Adaptive thermostats for noisy gradient systems
- Convergence rates for nonequilibrium Langevin dynamics
- Couplings and quantitative contraction rates for Langevin dynamics
- Ergodic Properties of Markov Processes
- Ergodicity for SDEs and approximations: locally Lipschitz vector fields and degenerate noise.
- Exponential relaxation of the Nosé-Hoover thermostat under Brownian heating
- scientific article; zbMATH DE number 1851000 (Why is no real title available?)
- Hypocoercivity
- Hypocoercivity for kinetic equations with linear relaxation terms
- Hypocoercivity for linear kinetic equations conserving mass
- Hypoelliptic second order differential equations
- Long-run accuracy of variational integrators in the stochastic context
- Markov chains and stochastic stability
- Molecular dynamics. With deterministic and stochastic numerical methods
- On the functional central limit theorem and the law of the iterated logarithm for Markov processes
- Partial differential equations and stochastic methods in molecular dynamics
- Recurrence and invariant measures for degenerate diffusions
- Reflection coupling and Wasserstein contractivity without convexity
- Spectral methods for Langevin dynamics and associated error estimates
- Stability of Markovian processes II: continuous-time processes and sampled chains
- The computation of averages from equilibrium and nonequilibrium Langevin molecular dynamics
- Theoretical and numerical comparison of some sampling methods for molecular dynamics
- Weak backward error analysis for Langevin process
- Yet Another Look at Harris’ Ergodic Theorem for Markov Chains
Cited in
(18)- Estimating the speed-up of adaptively restrained Langevin dynamics
- Hypocoercivity with Schur complements
- A functional law of the iterated logarithm for weakly hypoelliptic diffusions at time zero
- The method of stochastic characteristics for linear second-order hypoelliptic equations
- Adaptive thermostats for noisy gradient systems
- Langevin dynamics with general kinetic energies
- Accurate and efficient splitting methods for dissipative particle dynamics
- Hilbert space hypocoercivity for the Langevin dynamics revisited
- Random Batch Sum-of-Gaussians Method for Molecular Dynamics Simulations of Particle Systems
- On explicit L^2-convergence rate estimate for underdamped Langevin dynamics
- Friction-adaptive descent: a family of dynamics-based optimization methods
- Ergodicity and long-time behavior of the Random Batch Method for interacting particle systems
- Generalized Bayesian likelihood-free inference
- Adaptive stepsize algorithms for Langevin dynamics
- Appropriate state-dependent friction coefficient accelerates kinetic Langevin dynamics
- Hypocoercivity meets lifts
- How to construct explicit decay rates for kinetic Fokker-Planck equations?
- Adaptive solution of the master equation in low dimensions
This page was built for publication: Hypocoercivity properties of adaptive Langevin dynamics
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5110577)
