Adaptive Gibbs samplers
Computation
2010-01-19 v1
Abstract
We consider various versions of adaptive Gibbs and Metropolis within-Gibbs samplers, which update their selection probabilities (and perhaps also their proposal distributions) on the fly during a run, by learning as they go in an attempt to optimise the algorithm. We present a cautionary example of how even a simple-seeming adaptive Gibbs sampler may fail to converge. We then present various positive results guaranteeing convergence of adaptive Gibbs samplers under certain conditions.
Cite
@article{arxiv.1001.2797,
title = {Adaptive Gibbs samplers},
author = {Krzysztof Latuszynski and Jeffrey S. Rosenthal},
journal= {arXiv preprint arXiv:1001.2797},
year = {2010}
}