Titre :

On Margins and Derandomisation in PAC-Bayes

Auteur(s) :

Biggs, Felix [Auteur]
The Inria London Programme [Inria-London]
Department of Computer science [University College of London] [UCL-CS]
Guedj, Benjamin [Auteur]
The Inria London Programme [Inria-London]
MOdel for Data Analysis and Learning [MODAL]
Inria-CWI [Inria-CWI]
Department of Computer science [University College of London] [UCL-CS]

Titre de la manifestation scientifique :

AISTATS 2022 - 25th International Conference on Artificial Intelligence and Statistics

Ville :

Valencia

Pays :

Espagne

Date de début de la manifestation scientifique :

2022-03-28

Discipline(s) HAL :

Informatique [cs]/Apprentissage [cs.LG]
Statistiques [stat]/Machine Learning [stat.ML]
Statistiques [stat]/Théorie [stat.TH]

Résumé en anglais : [en]

We give a general recipe for derandomising PAC-Bayesian bounds using margins, with the critical ingredient being that our randomised predictions concentrate around some value. The tools we develop traightforwardly lead to ...
Lire la suite >We give a general recipe for derandomising PAC-Bayesian bounds using margins, with the critical ingredient being that our randomised predictions concentrate around some value. The tools we develop traightforwardly lead to margin bounds for various classifiers, including linear prediction—a class that includes boosting and the support vector machine—single-hidden-layer neural networks with an unusual erf activation function, and deep ReLU networks. Further, we extend to partially-derandomised predictors where only some of the randomness is removed, letting us extend bounds to cases where the concentration properties of our predictors are otherwise poor.Lire moins >

Langue :

Anglais

Comité de lecture :

Oui

Audience :

Internationale

Vulgarisation :

Non

Projet ANR :

Apprentissage PAC-bayésien agnostique
Une Perspective PAC-Bayésienne de l'Apprentissage de Représentations

Collections :

• Laboratoire Paul Painlevé - UMR 8524

Source :

Harvested from HAL