A function space HMC algorithm with second order Langevin diffusion limit

DOI10.3150/14-BEJ621zbMath1346.60119arXiv1308.0543OpenAlexW1872439012MaRDI QIDQ5963495

Andrew M. Stuart, Natesh S. Pillai, Michela Ottobre, Frank J. Pinski

Publication date: 22 February 2016

Published in: Bernoulli (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1308.0543

zbMATH Keywords

diffusion limitshybrid Monte Carlo algorithmMarkov chain Monte Carlo algorithmsecond-order Langevin diffusion

Mathematics Subject Classification ID

Computational methods in Markov chains (60J22) Central limit and other weak theorems (60F05) Monte Carlo methods (65C05) Markov chains (discrete-time Markov processes on discrete state spaces) (60J10) Numerical analysis or methods applied to Markov chains (65C40) Diffusion processes (60J60)

Related Items

Continuum limit and preconditioned Langevin sampling of the path integral molecular dynamics ⋮ FEM-based discretization-invariant MCMC methods for PDE-constrained Bayesian inverse problems ⋮ Markov chain Monte Carlo and irreversibility ⋮ On Irreversible Metropolis Sampling Related to Langevin Dynamics ⋮ Exact targeting of Gibbs distributions using velocity-jump processes ⋮ Advanced Multilevel Monte Carlo Methods ⋮ Reversible and non-reversible Markov chain Monte Carlo algorithms for reservoir simulation problems ⋮ Non-reversible guided Metropolis kernel ⋮ Analysis of a Computational Framework for Bayesian Inverse Problems: Ensemble Kalman Updates and MAP Estimators under Mesh Refinement ⋮ Optimal scaling of the MALA algorithm with irreversible proposals for Gaussian targets ⋮ Hamiltonian-Assisted Metropolis Sampling ⋮ Using perturbed underdamped Langevin dynamics to efficiently sample from probability distributions ⋮ MALA-within-Gibbs Samplers for High-Dimensional Distributions with Sparse Conditional Structure ⋮ High-dimensional MCMC with a standard splitting scheme for the underdamped Langevin diffusion ⋮ Is there an analog of Nesterov acceleration for gradient-based MCMC? ⋮ Modified Hamiltonian Monte Carlo for Bayesian inference ⋮ Peskun-Tierney ordering for Markovian Monte Carlo: beyond the reversible scenario ⋮ Limit theorems for the zig-zag process ⋮ Two Metropolis--Hastings Algorithms for Posterior Measures with Non-Gaussian Priors in Infinite Dimensions ⋮ On the geometric ergodicity of Hamiltonian Monte Carlo ⋮ Irreversible samplers from jump and continuous Markov processes ⋮ Mixing rates for Hamiltonian Monte Carlo algorithms in finite and infinite dimensions ⋮ Diffusion limit for the random walk Metropolis algorithm out of stationarity ⋮ Geometric Integration of Measure-Preserving Flows for Sampling ⋮ Non-reversible processes: GENERIC, hypocoercivity and fluctuations

Cites Work

• A patch that imparts unconditional stability to explicit integrators for Langevin-like equations
• Diffusion limits of the random walk Metropolis algorithm in high dimensions
• Noisy gradient flow from a random walk in Hilbert space
• Accelerating diffusions
• Hybrid Monte Carlo on Hilbert spaces
• Asymptotic behaviour of a class of stochastic approximation procedures
• Accelerating Gaussian diffusions
• Analysis of a nonreversible Markov chain sampler.
• Optimal scaling and diffusion limits for the Langevin algorithm in high dimensions
• A concise course on stochastic partial differential equations
• Invariant measure of the stochastic Allen-Cahn equation: the regime of small noise and large system size
• Inverse problems: A Bayesian perspective
• MCMC METHODS FOR DIFFUSION BRIDGES
• Pathwise accuracy and ergodicity of metropolized integrators for SDEs
• Monte Carlo sampling methods using Markov chains and their applications
• MCMC methods for functions: modifying old algorithms to make them faster
• Unnamed Item
• Unnamed Item
• Unnamed Item