Pathwise Accuracy and Ergodicity of Metropolized Integrators for SDEs

Abstract

Metropolized integrators for ergodic stochastic differential equations (SDE) are proposed which (i) are ergodic with respect to the (known) equilibrium distribution of the SDE and (ii) approximate pathwise the solutions of the SDE on finite time intervals. Both these properties are demonstrated in the paper and precise strong error estimates are obtained. It is also shown that the Metropolized integrator retains these properties even in situations where the drift in the SDE is nonglobally Lipschitz, and vanilla explicit integrators for SDEs typically become unstable and fail to be ergodic.

46 pages, 5 figures

References in corpus (1)

• Accurate sampling using Langevin dynamics

Cited by in corpus (21)

• Robust and efficient configurational molecular sampling via Langevin Dynamics
• The computation of averages from equilibrium and nonequilibrium Langevin molecular dynamics
• Convergence of Numerical Time-Averaging and Stationary Measures via Poisson Equations
• Diffusion limits of the random walk Metropolis algorithm in high dimensions
• Geometric integrators and the Hamiltonian Monte Carlo method
• Temporal Integrators for Fluctuating Hydrodynamics
• Non-intrusive and structure preserving multiscale integration of stiff ODEs, SDEs and Hamiltonian systems with hidden slow dynamics via flow averaging
• Deterministic Mean-field Ensemble Kalman Filtering
• Dimension-free path-integral molecular dynamics without preconditioning
• A Paradox of State-Dependent Diffusion and How to Resolve It
• Using nonequilibrium fluctuation theorems to understand and correct errors in equilibrium and nonequilibrium discrete Langevin dynamics simulations
• Error bounds for Metropolis-Hastings algorithms applied to perturbations of Gaussian measures in high dimensions
• A fast spectral method for electrostatics in doubly-periodic slit channels
• Metropolis Integration Schemes for Self-Adjoint Diffusions
• Langevin dynamics with general kinetic energies
• Numerical Unique Ergodicity of Monotone SDEs driven by Nondegenerate Multiplicative Noise
• Measuring the Irreversibility of Numerical Schemes for Reversible Stochastic Differential Equations
• Mixing of Metropolis-Adjusted Markov Chains via Couplings: The High Acceptance Regime
• Numerical Ergodicity and Uniform Estimate of Monotone SPDEs Driven by Multiplicative Noise
• Randomized Runge-Kutta-Nyström Methods for Unadjusted Hamiltonian and Kinetic Langevin Monte Carlo
• An Unstructured Mesh Reaction-Drift-Diffusion Master Equation with Reversible Reactions