Is the NUTS algorithm correct?
Computation
2020-05-08 v2 Numerical Analysis
Numerical Analysis
Abstract
This paper is devoted to investigate whether the popular No U-turn (NUTS) sampling algorithm is correct, i.e.\ whether the target probability distribution is \emph{exactly} conserved by the algorithm. It turns out that one of the Gibbs substeps used in the algorithm cannot always be guaranteed to be correct.
Cite
@article{arxiv.2005.01336,
title = {Is the NUTS algorithm correct?},
author = {J. M. Sanz-Serna},
journal= {arXiv preprint arXiv:2005.01336},
year = {2020}
}
Comments
Some statements in the paper are misleading. It is possible to think of NUTS at not being a slice/Gibbs sampler and, with an alternative interpretation, it may be be possible to prove that the algorithm is correct. In addition the experiment reported in Figure 2 should have had many initial states drawn from the target rather than using a single value