Split-and-Augmented Gibbs Sampler—Application to Large-Scale Inference Problems

DOI10.1109/TSP.2019.2894825MaRDI QIDQ4628330zbMATH OpenOpenAlexFDO

Authors Maxime Vono, Nicolas Dobigeon, Pierre Chainais

Publication date 6 March 2019

Published in IEEE Transactions on Signal Processing (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1804.05809

Mathematics Subject Classification ID

Monte Carlo methods (65C05) Analysis of algorithms (68W40) Applications of mathematical programming (90C90) Signal theory (characterization, reconstruction, filtering, etc.) (94A12) Sampling theory in information and communication theory (94A20)

Abstract: This paper derives two new optimization-driven Monte Carlo algorithms inspired from variable splitting and data augmentation. In particular, the formulation of one of the proposed approaches is closely related to the alternating direction method of multipliers (ADMM) main steps. The proposed framework enables to derive faster and more efficient sampling schemes than the current state-of-the-art methods and can embed the latter. By sampling efficiently the parameter to infer as well as the hyperparameters of the problem, the generated samples can be used to approximate Bayesian estimators of the parameters to infer. Additionally, the proposed approach brings confidence intervals at a low cost contrary to optimization methods. Simulations on two often-studied signal processing problems illustrate the performance of the two proposed samplers. All results are compared to those obtained by recent state-of-the-art optimization and MCMC algorithms used to solve these problems.

Cited in

(16)

• Accelerating Proximal Markov Chain Monte Carlo by Using an Explicit Stabilized Method
• High-dimensional Gaussian sampling: a review and a unifying approach based on a stochastic proximal point algorithm
• A proximal Markov chain Monte Carlo method for Bayesian inference in imaging inverse problems: when Langevin meets Moreau
• Global Consensus Monte Carlo
• Asymptotically Exact Data Augmentation: Models, Properties, and Algorithms
• New models for symbolic data analysis
• Approximate blocked Gibbs sampling for Bayesian neural networks
• The Split Gibbs Sampler Revisited: Improvements to Its Algorithmic Structure and Augmented Target Distribution
• Efficient Bayesian Computation for Low-Photon Imaging Problems
• Emerging directions in Bayesian computation
• Accelerated Bayesian imaging by relaxed proximal-point Langevin sampling
• Subgradient Langevin methods for sampling from nonsmooth potentials
• A Distributed Block-Split Gibbs Sampler with Hypergraph Structure for High-Dimensional Inverse Problems
• Sampling strategies in Bayesian inversion: a study of RTO and Langevin methods
• A contraction theory for Sinkhorn and Schrödinger bridges via log-Sobolev inequalities
• On the contraction properties of Sinkhorn semigroups

This page was built for publication: Split-and-Augmented Gibbs Sampler—Application to Large-Scale Inference Problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4628330)