Double descent for least-squares interpolation on contaminated data: A simulation study

TL;DR

This simulation study investigates the double descent phenomenon in overparametrized linear regression models with contaminated data, showing that least-squares interpolation can outperform robust methods in generalization.

Contribution

It demonstrates that double descent occurs in contaminated data settings and that least-squares interpolation can outperform robust alternatives in overparametrized linear regression.

Findings

Double descent observed in contaminated data with overparametrized models.

Least-squares interpolation surpasses robust methods in generalization performance.

Overparametrization enables good generalization despite data contamination.

Abstract

Overparametrized models can exhibit an excellent generalization performance, although they should be prone to overfitting according to classical statistical theory. The discovery of the "double descent", indicating that the generalization error decreases after a certain model complexity has been reached, opened a new line of research. Robust statistics considers statistical estimation on contaminated data, which, due to assumptions that do not hold on real data, let data points appear as outliers w.r.t. the assumed "ideal" distribution, potentially severely distorting any classical estimator. We address the question whether a double descent phenomenon can be observed in a linear regression setting with contaminated training data. We compare the performance of the highly non-robust least-squares interpolation estimator with several robust alternatives. It turns out that large overparametrization indeed allows for a double descent phenomenon, resulting in a very good generalization performance of the least-squares interpolator, surpassing that of the robust alternatives.