Robust Label Shift Quantification
Statistics Theory
2026-02-12 v2 Machine Learning
Statistics Theory
Abstract
In this paper, we investigate the label shift quantification problem. We propose robust estimators of the label distribution which turn out to coincide with the Maximum Likelihood Estimator. We analyze the theoretical aspects and derive deviation bounds for the proposed method, providing optimal guarantees in the well-specified case, along with notable robustness properties against outliers and contamination. Our results provide theoretical validation for empirical observations on the robustness of Maximum Likelihood Label Shift.
Keywords
Cite
@article{arxiv.2502.03174,
title = {Robust Label Shift Quantification},
author = {Alexandre Lecestre},
journal= {arXiv preprint arXiv:2502.03174},
year = {2026}
}
Comments
Revision were made, including a change of title. Also, this version contains new results in the calibration section