MONK -- Outlier-Robust Mean Embedding Estimation by Median-of-Means

Abstract : Mean embeddings provide an extremely flexible and powerful tool in machine learning and statistics to represent probability distributions and define a semi-metric (MMD, maximum mean discrepancy; also called N-distance or energy distance), with numerous successful applications. The representation is constructed as the expectation of the feature map defined by a kernel. As a mean, its classical empirical estimator, however, can be arbitrary severely affected even by a single outlier in case of unbounded features. To the best of our knowledge, unfortunately even the consistency of the existing few techniques trying to alleviate this serious sensitivity bottleneck is unknown. In this paper, we show how the recently emerged principle of median-of-means can be used to design estimators for kernel mean embedding and MMD with excessive resistance properties to outliers, and optimal sub-Gaussian deviation bounds under mild assumptions.
Type de document :
Rapport
[Research Report] CNRS / LMO - Laboratoire de Mathématiques d'Orsay, Orsay; Ecole Polytechnique (Palaiseau, France); Laboratoire de Mathématiques d'Orsay; ENSAE ParisTech; INRIA Saclay, équipe SELECT. 2018
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01705881
Contributeur : Zoltan Szabo <>
Soumis le : mercredi 17 octobre 2018 - 19:30:04
Dernière modification le : jeudi 15 novembre 2018 - 01:20:06

Fichier

MONK_TR.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01705881, version 4

Citation

Matthieu Lerasle, Zoltán Szabó, Timothée Mathieu, Guillaume Lecué. MONK -- Outlier-Robust Mean Embedding Estimation by Median-of-Means. [Research Report] CNRS / LMO - Laboratoire de Mathématiques d'Orsay, Orsay; Ecole Polytechnique (Palaiseau, France); Laboratoire de Mathématiques d'Orsay; ENSAE ParisTech; INRIA Saclay, équipe SELECT. 2018. 〈hal-01705881v4〉

Partager

Métriques

Consultations de la notice

18

Téléchargements de fichiers

4