Mixing times of data-augmentation Gibbs samplers for high-dimensional probit regression

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ascolani, Filippo, Zanella, Giacomo
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913848811847680
author Ascolani, Filippo
Zanella, Giacomo
author_facet Ascolani, Filippo
Zanella, Giacomo
contents We investigate the convergence properties of popular data-augmentation samplers for Bayesian probit regression. Leveraging recent results on Gibbs samplers for log-concave targets, we provide simple and explicit non-asymptotic bounds on the associated mixing times (in Kullback-Leibler divergence). The bounds depend explicitly on the design matrix and the prior precision, while they hold uniformly over the vector of responses. We specialize the results for different regimes of statistical interest, when both the number of data points $n$ and parameters $p$ are large: in particular we identify scenarios where the mixing times remain bounded as $n,p\to\infty$, and ones where they do not. The results are shown to be tight (in the worst case with respect to the responses) and provide guidance on choices of prior distributions that provably lead to fast mixing. An empirical analysis based on coupling techniques suggests that the bounds are effective in predicting practically observed behaviours.
format Preprint
id arxiv_https___arxiv_org_abs_2505_14343
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Mixing times of data-augmentation Gibbs samplers for high-dimensional probit regression
Ascolani, Filippo
Zanella, Giacomo
Computation
Methodology
Machine Learning
We investigate the convergence properties of popular data-augmentation samplers for Bayesian probit regression. Leveraging recent results on Gibbs samplers for log-concave targets, we provide simple and explicit non-asymptotic bounds on the associated mixing times (in Kullback-Leibler divergence). The bounds depend explicitly on the design matrix and the prior precision, while they hold uniformly over the vector of responses. We specialize the results for different regimes of statistical interest, when both the number of data points $n$ and parameters $p$ are large: in particular we identify scenarios where the mixing times remain bounded as $n,p\to\infty$, and ones where they do not. The results are shown to be tight (in the worst case with respect to the responses) and provide guidance on choices of prior distributions that provably lead to fast mixing. An empirical analysis based on coupling techniques suggests that the bounds are effective in predicting practically observed behaviours.
title Mixing times of data-augmentation Gibbs samplers for high-dimensional probit regression
topic Computation
Methodology
Machine Learning
url https://arxiv.org/abs/2505.14343