U-TraCE: a conformal prediction approach to uncertainty quantification in black-box models
Machine Learning: Science and Technology · 01.02.2026 · 7 (1) · 015017
DOI: 10.1088/2632-2153/ae35ce ↗Kurz gefasst
Fachartikel in der Zeitschrift Machine Learning: Science and Technology, online erschienen am 21.01.2026. Verfasst von Edgardo J Marchi und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Modern machine learning (ML) systems, particularly deep neural networks, have become increasingly complex and opaque, making reliable uncertainty quantification (UQ) essential. Conventional UQ methods often prove inadequate for black-box model evaluation scenarios, either because model internals are not accessible or because the resulting uncertainty is expressed in terms of a related predictor rather than the model under evaluation. We introduce U-TraCE (Uncertainty Tracking for Complex Estimators), a novel UQ method built on the Conformal Prediction (CP) framework that provides formal and traceable uncertainty bounds for opaque ML systems. Unlike existing CP approaches, which describe uncertainty through confidence intervals or prediction set sizes at a fixed uncertainty level, U-TraCE yields an explicit numerical bound on the error probability of the standalone black-box model. U-TraCE was evaluated on diverse tasks, including MNIST classification and cardiac magnetic resonance imaging (MRI) segmentation, representing both benchmark and real-world domains. Across these settings, it produced conservative yet well-calibrated uncertainty bounds that closely follow empirical uncertainty, even with limited calibration data. By bridging theoretical UQ and operational model validation, U-TraCE establishes a foundation for objective and automated conformity assessment of complex ML systems, supporting trustworthy deployment in safety-critical environments.
Zitieren
APA
Marchi, E. J., & Liebl, M. (2026). U-TraCE: a conformal prediction approach to uncertainty quantification in black-box models. Machine Learning: Science and Technology, 7(1), 015017. https://doi.org/10.1088/2632-2153/ae35ce
Vancouver
Marchi EJ, Liebl M. U-TraCE: a conformal prediction approach to uncertainty quantification in black-box models. Machine Learning: Science and Technology. 2026;7(1):015017. doi:10.1088/2632-2153/ae35ce
BibTeX
@article{Marchi2026,
author = {Marchi, Edgardo J and Liebl, Maik},
title = {U-TraCE: a conformal prediction approach to uncertainty quantification in black-box models},
journal = {Machine Learning: Science and Technology},
year = {2026},
volume = {7},
number = {1},
pages = {015017},
doi = {10.1088/2632-2153/ae35ce}
}
EUVIMED