Measuring Classification Decision Certainty and Doubt
Machine Learning
2023-03-29 v2 Artificial Intelligence
Machine Learning
Differential Geometry
Probability
Abstract
Quantitative characterizations and estimations of uncertainty are of fundamental importance in optimization and decision-making processes. Herein, we propose intuitive scores, which we call certainty and doubt, that can be used in both a Bayesian and frequentist framework to assess and compare the quality and uncertainty of predictions in (multi-)classification decision machine learning problems.
Keywords
Cite
@article{arxiv.2303.14568,
title = {Measuring Classification Decision Certainty and Doubt},
author = {Alexander M. Berenbeim and Iain J. Cruickshank and Susmit Jha and Robert H. Thomson and Nathaniel D. Bastian},
journal= {arXiv preprint arXiv:2303.14568},
year = {2023}
}
Comments
4 pages