On an adaptive preconditioned Crank-Nicolson MCMC algorithm for infinite dimensional Bayesian inference
From MaRDI portal
Publication:680134
DOI10.1016/J.JCP.2016.11.024zbMATH Open1380.65011arXiv1511.05838OpenAlexW2964057283MaRDI QIDQ680134FDOQ680134
Authors: Zixi Hu, Zhewei Yao, Jinglai Li 
Publication date: 22 January 2018
Published in: Journal of Computational Physics (Search for Journal in Brave)
Abstract: Many scientific and engineering problems require to perform Bayesian inferences for unknowns of infinite dimension. In such problems, many standard Markov Chain Monte Carlo (MCMC) algorithms become arbitrary slow under the mesh refinement, which is referred to as being dimension dependent. To this end, a family of dimensional independent MCMC algorithms, known as the preconditioned Crank-Nicolson (pCN) methods, were proposed to sample the infinite dimensional parameters. In this work we develop an adaptive version of the pCN algorithm, where the covariance operator of the proposal distribution is adjusted based on sampling history to improve the simulation efficiency. We show that the proposed algorithm satisfies an important ergodicity condition under some mild assumptions. Finally we provide numerical examples to demonstrate the performance of the proposed method.
Full work available at URL: https://arxiv.org/abs/1511.05838
Recommendations
- On a generalization of the preconditioned Crank-Nicolson metropolis algorithm
- Efficient Adaptive MCMC Through Precision Estimation
- Adaptive Bayesian inference on the mean of an infinite-dimensional normal distribution
- An Adaptive Independence Sampler MCMC Algorithm for Bayesian Inferences of Functions
- An adaptive approach to Langevin MCMC
- On the efficiency of adaptive MCMC algorithms
- An adaptive sequential Monte Carlo method for approximate Bayesian computation
- Two-scale coupling for preconditioned Hamiltonian Monte Carlo in infinite dimensions
- Optimal proposal distributions and adaptive MCMC
Bayesian inference (62F15) Monte Carlo methods (65C05) Inverse problems involving ordinary differential equations (34A55) Inverse problems for PDEs (35R30)
Cites Work
- Weak convergence and optimal scaling of random walk Metropolis algorithms
- Gaussian processes for machine learning.
- An adaptive Metropolis algorithm
- Statistical and computational inverse problems.
- Optimal scaling for various Metropolis-Hastings algorithms.
- Inverse problems: a Bayesian perspective
- An introduction to infinite-dimensional analysis
- Optimal scalings for local Metropolis-Hastings chains on nonproduct targets in high dimensions
- A stochastic Newton MCMC method for large-scale statistical inverse problems with application to seismic inversion
- MCMC methods for functions: modifying old algorithms to make them faster
- Proposals which speed up function-space MCMC
- Solving large-scale PDE-constrained Bayesian inverse problems with Riemann manifold Hamiltonian Monte Carlo
- A Bayesian inference approach to identify a Robin coefficient in one-dimensional parabolic problems
- Diffusion limits of the random walk Metropolis algorithm in high dimensions
- A computational framework for infinite-dimensional Bayesian inverse problems. II: stochastic Newton MCMC with application to ice sheet flow inverse problems
- On a generalization of the preconditioned Crank-Nicolson metropolis algorithm
- Dimension-independent likelihood-informed MCMC
- Accelerated dimension-independent adaptive metropolis
- A TV-Gaussian prior for infinite-dimensional Bayesian inverse problems and its numerical implementations
- Algorithms for Kullback-Leibler approximation of probability measures in infinite dimensions
- An Adaptive Independence Sampler MCMC Algorithm for Bayesian Inferences of Functions
Cited In (9)
- Bayesian updating and marginal likelihood estimation by cross entropy based importance sampling
- On a generalization of the preconditioned Crank-Nicolson metropolis algorithm
- Ergodicity of Markov chain Monte Carlo with reversible proposal
- Parallel MCMC algorithms: theoretical foundations, algorithm design, case studies
- Non-stationary multi-layered Gaussian priors for Bayesian inversion
- A hybrid adaptive MCMC algorithm in function spaces
- Ensemble sampler for infinite-dimensional inverse problems
- Geometric MCMC for infinite-dimensional inverse problems
- A numerical method for solving linear systems in the preconditioned Crank-Nicolson algorithm
This page was built for publication: On an adaptive preconditioned Crank-Nicolson MCMC algorithm for infinite dimensional Bayesian inference
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q680134)
