Accelerated Gibbs sampling of normal distributions using matrix splittings and polynomials
From MaRDI portal
Abstract: Standard Gibbs sampling applied to a multivariate normal distribution with a specified precision matrix is equivalent in fundamental ways to the Gauss-Seidel iterative solution of linear equations in the precision matrix. Specifically, the iteration operators, the conditions under which convergence occurs, and geometric convergence factors (and rates) are identical. These results hold for arbitrary matrix splittings from classical iterative methods in numerical linear algebra giving easy access to mature results in that field, including existing convergence results for antithetic-variable Gibbs sampling, REGS sampling, and generalizations. Hence, efficient deterministic stationary relaxation schemes lead to efficient generalizations of Gibbs sampling. The technique of polynomial acceleration that significantly improves the convergence rate of an iterative solver derived from a emph{symmetric} matrix splitting may be applied to accelerate the equivalent generalized Gibbs sampler. Identicality of error polynomials guarantees convergence of the inhomogeneous Markov chain, while equality of convergence factors ensures that the optimal solver leads to the optimal sampler. Numerical examples are presented, including a Chebyshev accelerated SSOR Gibbs sampler applied to a stylized demonstration of low-level Bayesian image reconstruction in a large 3-dimensional linear inverse problem.
Recommendations
- Convergence in variance of Chebyshev accelerated Gibbs samplers
- scientific article; zbMATH DE number 1085989
- Rate of convergence of the Gibbs sampler in the Gaussian case
- Use of the Gibbs sampler to invert large, possibly sparse, positive definite matrices
- scientific article; zbMATH DE number 549940
Cited in
(15)- Use of the Gibbs sampler to invert large, possibly sparse, positive definite matrices
- A new inherent reliability modeling and analysis method based on imprecise Dirichlet model for machine tool spindle
- Localization for MCMC: sampling high-dimensional posterior distributions with local structure
- Approximation and sampling of multivariate probability distributions in the tensor train decomposition
- Approximation algorithms for the normalizing constant of Gibbs distributions
- Factorization of separable and patterned covariance matrices for Gibbs sampling
- Polynomial accelerated MCMC and other sampling algorithms inspired by computational optimization
- Polynomial Accelerated Solutions to a Large Gaussian Model for Imaging Biofilms: In Theory and Finite Precision
- MALA-within-Gibbs samplers for high-dimensional distributions with sparse conditional structure
- High-dimensional Gaussian sampling: a review and a unifying approach based on a stochastic proximal point algorithm
- A blocking scheme for dimension-robust Gibbs sampling in large-scale image deblurring
- Rank bounds for approximating Gaussian densities in the tensor-train format
- Sampling hyperparameters in hierarchical models: Improving on Gibbs for high-dimensional latent fields and large datasets
- Convergence in variance of Chebyshev accelerated Gibbs samplers
- Uncertainty quantification. Abstracts from the workshop held April 20--25, 2025
This page was built for publication: Accelerated Gibbs sampling of normal distributions using matrix splittings and polynomials
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2405171)
