Differential Privacy has Bounded Impact on Fairness in Classification

Mangold, Paul; Perrot, Michaël; Bellet, Aurélien; Tommasi, Marc

Type de document :

Pré-publication ou Document de travail

Titre :

Differential Privacy has Bounded Impact on Fairness in Classification

Auteur(s) :

Mangold, Paul [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Machine Learning in Information Networks [MAGNET]
Perrot, Michaël [Auteur]
Machine Learning in Information Networks [MAGNET]
Bellet, Aurélien [Auteur]
Machine Learning in Information Networks [MAGNET]
Tommasi, Marc [Auteur]

Machine Learning in Information Networks [MAGNET]

Date de publication :

2022-10-28

Mot(s)-clé(s) en anglais :

Fairness
Differential privacy
Classification
Machine Learning

Discipline(s) HAL :

Informatique [cs]/Apprentissage [cs.LG]

Résumé en anglais : [en]

We theoretically study the impact of differential privacy on fairness in classification. We prove that, given a class of models, popular group fairness measures are pointwise Lipschitz-continuous with respect to the ...
Lire la suite >We theoretically study the impact of differential privacy on fairness in classification. We prove that, given a class of models, popular group fairness measures are pointwise Lipschitz-continuous with respect to the parameters of the model. This result is a consequence of a more general statement on accuracy conditioned on an arbitrary event (such as membership to a sensitive group), which may be of independent interest. We use the aforementioned Lipschitz property to prove a high probability bound showing that, given enough examples, the fairness level of private models is close to the one of their non-private counterparts.Lire moins >

Langue :

Anglais

Commentaire :

18 pages

Collections :