Regularized least squares learning with heavy-tailed noise is minimax optimal
By: Mattes Mollenhauer , Nicole Mücke , Dimitri Meunier and more
Potential Business Impact:
Makes computer learning work better with messy data.
This paper examines the performance of ridge regression in reproducing kernel Hilbert spaces in the presence of noise that exhibits a finite number of higher moments. We establish excess risk bounds consisting of subgaussian and polynomial terms based on the well known integral operator framework. The dominant subgaussian component allows to achieve convergence rates that have previously only been derived under subexponential noise - a prevalent assumption in related work from the last two decades. These rates are optimal under standard eigenvalue decay conditions, demonstrating the asymptotic robustness of regularized least squares against heavy-tailed noise. Our derivations are based on a Fuk-Nagaev inequality for Hilbert-space valued random variables.
Similar Papers
Understanding Robust Machine Learning for Nonparametric Regression with Heavy-Tailed Noise
Machine Learning (CS)
Makes computers learn from messy, unreliable data.
Minimax Optimal Robust Sparse Regression with Heavy-Tailed Designs: A Gradient-Based Approach
Methodology
Makes computer learning work with messy data.
Heavy Lasso: sparse penalized regression under heavy-tailed noise via data-augmented soft-thresholding
Methodology
Makes computer models better with messy data.