Stability of adversarial Markov chains, with an application to adaptive MCMC algorithms

DOI10.1214/14-AAP1083zbMATH Open1328.60169arXiv1403.3950MaRDI QIDQ894817FDOQ894817

Authors: Radu V. Craiu, Krzysztof Łatuszyński, Neal Madras, Gareth O. Roberts, Jeffrey S. Rosenthal, Lawrence F. Gray 

Publication date: 24 November 2015

Published in: The Annals of Applied Probability (Search for Journal in Brave)

Abstract: We consider whether ergodic Markov chains with bounded step size remain bounded in probability when their transitions are modified by an adversary on a bounded subset. We provide counterexamples to show that the answer is no in general, and prove theorems to show that the answer is yes under various additional assumptions. We then use our results to prove convergence of various adaptive Markov chain Monte Carlo algorithms.

Full work available at URL: https://arxiv.org/abs/1403.3950

Recommendations

• On the containment condition for adaptive Markov chain Monte Carlo algorithms
• On the ergodicity properties of some adaptive MCMC algorithms
• Convergence of adaptive and interacting Markov chain Monte Carlo algorithms
• On adaptive Markov chain Monte Carlo algorithms
• On the convergence rates of some adaptive Markov chain Monte Carlo algorithms
• Stability of Markov chain Monte Carlo methods (to appear)
• Coupling and Ergodicity of Adaptive Markov Chain Monte Carlo Algorithms
• Stability of sequential Markov Chain Monte Carlo methods
• Approximate adaptive uniformization of continuous-time Markov chains
• Ergodicity of combocontinuous adaptive MCMC algorithms

zbMATH Keywords

stabilityperturbationsadaptive Markov chain Monte Carlo algorithmsergodic Markov chains

Mathematics Subject Classification ID

Computational methods in Markov chains (60J22) Point estimation (62F10) Bayesian inference (62F15) Monte Carlo methods (65C05) Numerical analysis or methods applied to Markov chains (65C40) Discrete-time Markov processes on general state spaces (60J05)

Cites Work

• Title not available (Why is that?)
• An adaptive Metropolis algorithm
• Handbook of Markov Chain Monte Carlo
• Coupling and Ergodicity of Adaptive Markov Chain Monte Carlo Algorithms
• Limit theorems for some adaptive MCMC algorithms with subgeometric kernels
• On the ergodicity of the adaptive Metropolis algorithm on unbounded domains
• Markov chains and stochastic stability
• On adaptive Markov chain Monte Carlo algorithms
• On the ergodicity properties of some adaptive MCMC algorithms
• General state space Markov chains and MCMC algorithms
• General Irreducible Markov Chains and Non-Negative Operators
• On the notion of recurrence in discrete stochastic processes
• Adaptive Gibbs samplers and related MCMC methods
• Optimal proposal distributions and adaptive MCMC
• Hitting-time and occupation-time bounds implied by drift analysis with applications
• Adversarial queuing theory
• Title not available (Why is that?)
• Moment conditions for a sequence with negative drift to be uniformly bounded in \(L^r\)
• On the containment condition for adaptive Markov chain Monte Carlo algorithms
• Learn from thy neighbor: parallel-chain and regional adaptive MCMC
• Two convergence properties of hybrid samplers
• The containment condition and AdapFail algorithms

Cited In (8)

• Adaptive Incremental Mixture Markov Chain Monte Carlo
• An adaptive multiple-try Metropolis algorithm
• Adaptive schemes for piecewise deterministic Monte Carlo algorithms
• Ergodicity of combocontinuous adaptive MCMC algorithms
• Accelerating adaptation in the adaptive Metropolis–Hastings random walk algorithm
• The containment condition and AdapFail algorithms
• Detecting Markov chain instability: a Monte Carlo approach
• Adaptive Component-Wise Multiple-Try Metropolis Sampling

Uses Software

• mcmcse

This page was built for publication: Stability of adversarial Markov chains, with an application to adaptive MCMC algorithms

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q894817)