Affordable Access

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

Authors
  • Lerasle, Matthieu
  • Szabó, Zoltán
  • Lecué, Guillaume
  • Massiot, Gaspar
  • Moulines, Eric
Publication Date
Feb 09, 2018
Source
HAL-Rennes 1
Keywords
Language
English
License
Unknown
External links

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 minimax-optimal estimators for kernel mean embedding and MMD, with finite-sample strong outlier-robustness guarantees.

Report this publication

Statistics

Seen <100 times