Uncertainty in Physics and AI: Taxonomy, Quantification, and Validation
Machine Learning
2026-05-12 v1 Cosmology and Nongalactic Astrophysics
Astrophysics of Galaxies
High Energy Physics - Experiment
High Energy Physics - Phenomenology
Abstract
Reliable uncertainty quantification is essential for the use of machine learning in physics, where scientific discoveries depend on validated probabilistic statements. We provide a structured overview of uncertainty quantification in ML for physics, introducing a unified taxonomy of uncertainty and clarifying the interpretation of predictive and inference uncertainties across frequentist and Bayesian frameworks. We discuss principled validation tools, including coverage, calibration, bias tests, and proper scoring rules, and illustrate them with simple regression and classification examples.
Cite
@article{arxiv.2605.10378,
title = {Uncertainty in Physics and AI: Taxonomy, Quantification, and Validation},
author = {Manuel Haußmann and Ramon Winterhalder and Maria Ubiali},
journal= {arXiv preprint arXiv:2605.10378},
year = {2026}
}