Simpler PAC-Bayesian Bounds for Hostile Data

Alquier, Pierre; Guedj, Benjamin

Document type :

Article dans une revue scientifique

DOI :

10.1007/s10994-017-5690-0

Permalink :

http://hdl.handle.net/20.500.12210/29360

Title :

Simpler PAC-Bayesian Bounds for Hostile Data

Author(s) :

Alquier, Pierre [Auteur]
Guedj, Benjamin [Auteur]

Laboratoire Paul Painlevé - UMR 8524 [LPP]

Journal title :

Machine Learning

Volume number :

107

Pages :

887–902

Publisher :

Springer Verlag

Publication date :

2018

ISSN :

0885-6125

Keyword(s) :

F-divergence
Oracle inequalities
Dependent and unbounded data
PAC-Bayesian theory

HAL domain(s) :

Statistiques [stat]/Machine Learning [stat.ML]

English abstract : [en]

PAC-Bayesian learning bounds are of the utmost interest to the learning community. Their role is to connect the generalization ability of an aggregation distribution $\\rho$ to its empirical risk and to its Kullback-Leibler ...
Show more >PAC-Bayesian learning bounds are of the utmost interest to the learning community. Their role is to connect the generalization ability of an aggregation distribution $\\rho$ to its empirical risk and to its Kullback-Leibler divergence with respect to some prior distribution $\\pi$. Unfortunately, most of the available bounds typically rely on heavy assumptions such as boundedness and independence of the observations. This paper aims at relaxing these constraints and provides PAC-Bayesian learning bounds that hold for dependent, heavy-tailed observations (hereafter referred to as \\emph{hostile data}). In these bounds the Kullack-Leibler divergence is replaced with a general version of Csisz\\'ar's $f$-divergence. We prove a general PAC-Bayesian bound, and show how to use it in various hostile settings.Show less >

Language :

Anglais

Audience :

Internationale

Popular science :

Non

Administrative institution(s) :

CNRS
Université de Lille

Collections :

METRICS : Evaluation des technologies de santé et des pratiques médicales - ULR 2694

Submission date :

2020-06-08T14:11:19Z
2020-06-09T09:17:15Z

Files

documen
Open access
Access the document

Version	Link	Modification date
2	20.500.12210/29360*	2020-06-09T09:17:00Z
1	20.500.12210/29360.1	2020-06-08T14:11:19Z

Simpler PAC-Bayesian Bounds for Hostile Data

BibTeX

CSV

Excel

RIS

Files

Version history

Simpler PAC-Bayesian Bounds for Hostile Data BibTeX CSV Excel RIS

Files

Version history

Simpler PAC-Bayesian Bounds for Hostile Data

BibTeX

CSV

Excel

RIS