A Statistical Learning Assessment of Huber Regression

TL;DR

This paper provides a theoretical assessment of Huber regression in nonparametric statistical learning, demonstrating its robustness and establishing convergence rates under weak moment conditions, with practical guidance on parameter tuning.

Contribution

It offers the first systematic theoretical analysis of Huber regression in nonparametric learning, highlighting the importance of adaptive scale tuning for robustness and mean regression consistency.

Findings

Risk consistency does not guarantee learnability in mean regression.

Adaptive tuning of the scale parameter is essential for effective Huber regression.

Convergence rates are established under $(1+\epsilon)$-moment conditions, including infinite variance cases.

Abstract

As one of the triumphs and milestones of robust statistics, Huber regression plays an important role in robust inference and estimation. It has also been finding a great variety of applications in machine learning. In a parametric setup, it has been extensively studied. However, in the statistical learning context where a function is typically learned in a nonparametric way, there is still a lack of theoretical understanding of how Huber regression estimators learn the conditional mean function and why it works in the absence of light-tailed noise assumptions. To address these fundamental questions, we conduct an assessment of Huber regression from a statistical learning viewpoint. First, we show that the usual risk consistency property of Huber regression estimators, which is usually pursued in machine learning, cannot guarantee their learnability in mean regression. Second, we argue that Huber regression should be implemented in an adaptive way to perform mean regression, implying that one needs to tune the scale parameter in accordance with the sample size and the moment condition of the noise. Third, with an adaptive choice of the scale parameter, we demonstrate that Huber regression estimators can be asymptotic mean regression calibrated under $(1+\epsilon)$-moment conditions ($\epsilon>0$). Last but not least, under the same moment conditions, we establish almost sure convergence rates for Huber regression estimators. Note that the $(1+\epsilon)$-moment conditions accommodate the special case where the response variable possesses infinite variance and so the established convergence rates justify the robustness feature of Huber regression estimators. In the above senses, the present study provides a systematic statistical learning assessment of Huber regression estimators and justifies their merits in terms of robustness from a theoretical viewpoint.