Time-uniform Chernoff bounds via nonnegative supermartingales

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Howard, Steven R., Ramdas, Aaditya, McAuliffe, Jon, Sekhon, Jasjeet
Format: Preprint
Published: 2018
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911323337523200
author Howard, Steven R.
Ramdas, Aaditya
McAuliffe, Jon
Sekhon, Jasjeet
author_facet Howard, Steven R.
Ramdas, Aaditya
McAuliffe, Jon
Sekhon, Jasjeet
contents We develop a class of exponential bounds for the probability that a martingale sequence crosses a time-dependent linear threshold. Our key insight is that it is both natural and fruitful to formulate exponential concentration inequalities in this way. We illustrate this point by presenting a single assumption and theorem that together unify and strengthen many tail bounds for martingales, including classical inequalities (1960-80) by Bernstein, Bennett, Hoeffding, and Freedman; contemporary inequalities (1980-2000) by Shorack and Wellner, Pinelis, Blackwell, van de Geer, and de la Peña; and several modern inequalities (post-2000) by Khan, Tropp, Bercu and Touati, Delyon, and others. In each of these cases, we give the strongest and most general statements to date, quantifying the time-uniform concentration of scalar, matrix, and Banach-space-valued martingales, under a variety of nonparametric assumptions in discrete and continuous time. In doing so, we bridge the gap between existing line-crossing inequalities, the sequential probability ratio test, the Cramér-Chernoff method, self-normalized processes, and other parts of the literature.
format Preprint
id arxiv_https___arxiv_org_abs_1808_03204
institution arXiv
publishDate 2018
record_format arxiv
spellingShingle Time-uniform Chernoff bounds via nonnegative supermartingales
Howard, Steven R.
Ramdas, Aaditya
McAuliffe, Jon
Sekhon, Jasjeet
Probability
60E15, 60G17 (Primary) 60F10, 60B20 (Secondary)
We develop a class of exponential bounds for the probability that a martingale sequence crosses a time-dependent linear threshold. Our key insight is that it is both natural and fruitful to formulate exponential concentration inequalities in this way. We illustrate this point by presenting a single assumption and theorem that together unify and strengthen many tail bounds for martingales, including classical inequalities (1960-80) by Bernstein, Bennett, Hoeffding, and Freedman; contemporary inequalities (1980-2000) by Shorack and Wellner, Pinelis, Blackwell, van de Geer, and de la Peña; and several modern inequalities (post-2000) by Khan, Tropp, Bercu and Touati, Delyon, and others. In each of these cases, we give the strongest and most general statements to date, quantifying the time-uniform concentration of scalar, matrix, and Banach-space-valued martingales, under a variety of nonparametric assumptions in discrete and continuous time. In doing so, we bridge the gap between existing line-crossing inequalities, the sequential probability ratio test, the Cramér-Chernoff method, self-normalized processes, and other parts of the literature.
title Time-uniform Chernoff bounds via nonnegative supermartingales
topic Probability
60E15, 60G17 (Primary) 60F10, 60B20 (Secondary)
url https://arxiv.org/abs/1808.03204