Deep-Learning Jets with Uncertainties and More
High Energy Physics - Phenomenology
2020-01-22 v2
Abstract
Bayesian neural networks allow us to keep track of uncertainties, for example in top tagging, by learning a tagger output together with an error band. We illustrate the main features of Bayesian versions of established deep-learning taggers. We show how they capture statistical uncertainties from finite training samples, systematics related to the jet energy scale, and stability issues through pile-up. Altogether, Bayesian networks offer many new handles to understand and control deep learning at the LHC without introducing a visible prior effect and without compromising the network performance.
Cite
@article{arxiv.1904.10004,
title = {Deep-Learning Jets with Uncertainties and More},
author = {Sven Bollweg and Manuel Haussmann and Gregor Kasieczka and Michel Luchmann and Tilman Plehn and Jennifer Thompson},
journal= {arXiv preprint arXiv:1904.10004},
year = {2020}
}
Comments
15 figures