A Vector Bernstein Inequality for Self-Normalized Martingales

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autore principale: Ziemann, Ingvar
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915090179031040
author Ziemann, Ingvar
author_facet Ziemann, Ingvar
contents We prove a Bernstein inequality for vector-valued self-normalized martingales. We first give an alternative perspective of the corresponding sub-Gaussian bound due to Abbasi-Yadkori et al. via a PAC-Bayesian argument with Gaussian priors. By instantiating this argument to priors drawn uniformly over well-chosen ellipsoids, we obtain a Bernstein bound.
format Preprint
id arxiv_https___arxiv_org_abs_2412_20949
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Vector Bernstein Inequality for Self-Normalized Martingales
Ziemann, Ingvar
Probability
We prove a Bernstein inequality for vector-valued self-normalized martingales. We first give an alternative perspective of the corresponding sub-Gaussian bound due to Abbasi-Yadkori et al. via a PAC-Bayesian argument with Gaussian priors. By instantiating this argument to priors drawn uniformly over well-chosen ellipsoids, we obtain a Bernstein bound.
title A Vector Bernstein Inequality for Self-Normalized Martingales
topic Probability
url https://arxiv.org/abs/2412.20949