Calibrating sufficiently
Machine Learning
2022-04-26 v5 Machine Learning
Statistics Theory
Statistics Theory
Abstract
When probabilistic classifiers are trained and calibrated, the so-called grouping loss component of the calibration loss can easily be overlooked. Grouping loss refers to the gap between observable information and information actually exploited in the calibration exercise. We investigate the relation between grouping loss and the concept of sufficiency, identifying comonotonicity as a useful criterion for sufficiency. We revisit the probing reduction approach of Langford & Zadrozny (2005) and find that it produces an estimator of probabilistic classifiers that reduces grouping loss. Finally, we discuss Brier curves as tools to support training and 'sufficient' calibration of probabilistic classifiers.
Keywords
Cite
@article{arxiv.2105.07283,
title = {Calibrating sufficiently},
author = {Dirk Tasche},
journal= {arXiv preprint arXiv:2105.07283},
year = {2022}
}
Comments
27 pages, 2 figures, appendix