Reversible and non-reversible Markov chain Monte Carlo algorithms for reservoir simulation problems

DOI10.1007/S10596-020-09947-4MaRDI QIDQ2192802zbMATH OpenOpenAlexFDO

Authors I. Fursov, M. Ottobre, Paul W. Dobson, Gabriel J. Lord

Publication date 18 August 2020

Published in Computational Geosciences (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/1903.06960

zbMATH Keywords

Markov chain Monte Carlo methodsnon-reversible Markov chainshigh-dimensional samplingsubsurface reservoir simulation

Mathematics Subject Classification ID

Monte Carlo methods (65C05) Potentials, prospecting (86A20)

Abstract: We compare numerically the performance of reversible and non-reversible Markov Chain Monte Carlo algorithms for high dimensional oil reservoir problems; because of the nature of the problem at hand, the target measures from which we sample are supported on bounded domains. We compare two strategies to deal with bounded domains, namely reflecting proposals off the boundary and rejecting them when they fall outside of the domain. We observe that for complex high dimensional problems reflection mechanisms outperform rejection approaches and that the advantage of introducing non-reversibility in the Markov Chain employed for sampling is more and more visible as the dimension of the parameter space increases.

Recommendations

• Analysis of a nonreversible Markov chain sampler.
• Markov chain Monte Carlo and irreversibility
• On the convergence time of some non-reversible Markov chain Monte Carlo methods
• The Bouncy Particle Sampler: A Non-Reversible Rejection-Free Markov Chain Monte Carlo Method
• Non-reversible Metropolis-Hastings

Cites work

• scientific article; zbMATH DE number 1243473 (Why is no real title available?)
• A function space HMC algorithm with second order Langevin diffusion limit
• Calculating derivatives for automatic history matching
• Free energy computations. A mathematical perspective
• Hybrid Monte Carlo on Hilbert spaces
• Hypocoercivity
• Irreversible Langevin samplers and variance reduction: a large deviations approach
• Markov chain Monte Carlo and irreversibility
• Markov chain Monte Carlo and numerical differential equations
• Piecewise deterministic Markov processes for scalable Monte Carlo on restricted domains
• The Bouncy Particle Sampler: A Non-Reversible Rejection-Free Markov Chain Monte Carlo Method
• Variance reduction for irreversible Langevin samplers and diffusion on graphs
• Variance reduction using nonreversible Langevin samplers

Cited in

(6)

• Efficient Bayesian Computation for Low-Photon Imaging Problems
• Reversible algorithm of simulating multivariate densities with multi-hump
• Non-reversible processes: GENERIC, hypocoercivity and fluctuations
• Geometric Integration of Measure-Preserving Flows for Sampling
• Reversibility violation in the hybrid Monte Carlo algorithm
• scientific article; zbMATH DE number 2222874 (Why is no real title available?)

Describes a project that uses

Uses Software

• ECLIPSE

This page was built for publication: Reversible and non-reversible Markov chain Monte Carlo algorithms for reservoir simulation problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2192802)