Foolish Crowds Support Benign Overfitting
Niladri S. Chatterji
Philip M. Long

Abstract
We prove a lower bound on the excess risk of sparse interpolating procedures for linear regression with Gaussian data in the overparameterized regime. We apply this result to obtain a lower bound for basis pursuit (the minimum -norm interpolant) that implies that its excess risk can converge at an exponentially slower rate than OLS (the minimum -norm interpolant), even when the ground truth is sparse. Our analysis exposes the benefit of an effect analogous to the "wisdom of the crowd", except here the harm arising from fitting the is ameliorated by spreading it among many directions -- the variance reduction arises from a crowd.
View on arXivComments on this paper