Skip to main content

Research Repository

Advanced Search

Combine and conquer: event reconstruction with Bayesian Ensemble Neural Networks

Araz, Jack Y.; Spannowsky, Michael

Combine and conquer: event reconstruction with Bayesian Ensemble Neural Networks Thumbnail


Authors



Abstract

Ensemble learning is a technique where multiple component learners are combined through a protocol. We propose an Ensemble Neural Network (ENN) that uses the combined latent-feature space of multiple neural network classifiers to improve the representation of the network hypothesis. We apply this approach to construct an ENN from Convolutional and Recurrent Neural Networks to discriminate top-quark jets from QCD jets. Such ENN provides the flexibility to improve the classification beyond simple prediction combining methods by linking different sources of error correlations, hence improving the representation between data and hypothesis. In combination with Bayesian techniques, we show that it can reduce epistemic uncertainties and the entropy of the hypothesis by simultaneously exploiting various kinematic correlations of the system, which also makes the network less susceptible to a limitation in training sample size.

Citation

Araz, J. Y., & Spannowsky, M. (2021). Combine and conquer: event reconstruction with Bayesian Ensemble Neural Networks. Journal of High Energy Physics, 2021(4), Article 296. https://doi.org/10.1007/jhep04%282021%29296

Journal Article Type Article
Acceptance Date Apr 2, 2021
Online Publication Date Apr 30, 2021
Publication Date 2021
Deposit Date Jul 28, 2021
Publicly Available Date Mar 28, 2024
Journal Journal of High Energy Physics
Print ISSN 1126-6708
Publisher Scuola Internazionale Superiore di Studi Avanzati (SISSA)
Peer Reviewed Peer Reviewed
Volume 2021
Issue 4
Article Number 296
DOI https://doi.org/10.1007/jhep04%282021%29296

Files

Published Journal Article (4.5 Mb)
PDF

Publisher Licence URL
http://creativecommons.org/licenses/by/4.0/

Copyright Statement
Open Access. This article is distributed under the terms of the Creative Commons Attribution License (CC-BY 4.0), which permits any use, distribution and reproduction in any medium, provided the original author(s) and source are credited.





You might also like



Downloadable Citations