Universal Robust Regression via Maximum Mean Discrepancy

TL;DR

This paper introduces a novel, model-agnostic robust regression method based on Maximum Mean Discrepancy, effective against outliers and adversarial contamination, with theoretical guarantees and new insights into kernel mean embeddings.

Contribution

It develops a universal robust regression approach applicable to any model using MMD minimization, with two estimators proven to handle Huber-type and adversarial contamination.

Findings

Two robust estimators with theoretical error bounds

One estimator is computationally efficient, the other more robust

New results on kernel conditional mean embedding

Abstract

Many modern datasets are collected automatically and are thus easily contaminated by outliers. This led to a regain of interest in robust estimation, including new notions of robustness such as robustness to adversarial contamination of the data. However, most robust estimation methods are designed for a specific model. Notably, many methods were proposed recently to obtain robust estimators in linear models (or generalized linear models), and a few were developed for very specific settings, for example beta regression or sample selection models. In this paper we develop a new approach for robust estimation in arbitrary regression models, based on Maximum Mean Discrepancy minimization. We build two estimators which are both proven to be robust to Huber-type contamination. We obtain a non-asymptotic error bound for one them and show that it is also robust to adversarial contamination, but this estimator is computationally more expensive to use in practice than the other one. As a by-product of our theoretical analysis of the proposed estimators we derive new results on kernel conditional mean embedding of distributions which are of independent interest.