Trimmed Maximum Likelihood Estimation for Robust Generalized Linear ModelDownload PDF

Published: 31 Oct 2022, Last Modified: 11 Jan 2023NeurIPS 2022 AcceptReaders: Everyone
Abstract: We study the problem of learning generalized linear models under adversarial corruptions. We analyze a classical heuristic called the \textit{iterative trimmed maximum likelihood estimator} which is known to be effective against \textit{label corruptions} in practice. Under label corruptions, we prove that this simple estimator achieves minimax near-optimal risk on a wide range of generalized linear models, including Gaussian regression, Poisson regression and Binomial regression. Finally, we extend the estimator to the much more challenging setting of \textit{label and covariate corruptions} and demonstrate its robustness and optimality in that setting as well.
Supplementary Material: pdf
15 Replies