Robust Estimation under the Wasserstein Distance

arXiv2302.01237MaRDI QIDQ6425257

Author name not available (Why is that?)

Publication date: 2 February 2023

Abstract: We study the problem of robust distribution estimation under the Wasserstein metric, a popular discrepancy measure between probability distributions rooted in optimal transport (OT) theory. We introduce a new outlier-robust Wasserstein distance $mathsf{W}_p^{v}arepsilon$ which allows for $varepsilon$ outlier mass to be removed from its input distributions, and show that minimum distance estimation under $mathsf{W}_p^{v}arepsilon$ achieves minimax optimal robust estimation risk. Our analysis is rooted in several new results for partial OT, including an approximate triangle inequality, which may be of independent interest. To address computational tractability, we derive a dual formulation for $mathsf{W}_p^{v}arepsilon$ that adds a simple penalty term to the classic Kantorovich dual objective. As such, $mathsf{W}_p^{v}arepsilon$ can be implemented via an elementary modification to standard, duality-based OT solvers. Our results are extended to sliced OT, where distributions are projected onto low-dimensional subspaces, and applications to homogeneity and independence testing are explored. We illustrate the virtues of our framework via applications to generative modeling with contaminated datasets.

Has companion code repository: https://github.com/sbnietert/robust-ot

This page was built for publication: Robust Estimation under the Wasserstein Distance

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6425257)