CERN Accelerating science

Article
Report number	arXiv:2104.14659
Title	End-to-End Jet Classification of Boosted Top Quarks with CMS Open Data
Related title	End-to-end jet classification of boosted top quarks with the CMS open data
Author(s)	Andrews, Michael (Carnegie Mellon U.) ; Burkle, Bjorn (Brown U.) ; Chen, Yi-fan (Digital Pathways, Mtn. View) ; DiCroce, Davide (Alabama U.) ; Gleyzer, Sergei (Alabama U.) ; Heintz, Ulrich (Brown U.) ; Narain, Meenakshi (Brown U.) ; Paulini, Manfred (Carnegie Mellon U.) ; Pervan, Nikolas (Brown U.) ; Shafi, Yusef (Google Inc. ; Digital Pathways, Mtn. View) ; Sun, Wei (Google Inc. ; Digital Pathways, Mtn. View) ; Usai, Emanuele (Brown U.) ; Yang, Kun (Google Inc. ; Digital Pathways, Mtn. View)
Publication	2021
Imprint	2021-04-19
Number of pages	9
Note	9 pages, 3 figures, 4 tables; v3: unpublished
In:	EPJ Web Conf. 251 (2021) pp.04030 Phys. Rev. D 105, 5 (2022) pp.052008
In:	25th International Conference on Computing in High-Energy and Nuclear Physics (CHEP), Online, Online, 17 - 21 May 2021, pp.04030
DOI	10.1051/epjconf/202125104030 (publication) 10.1103/PhysRevD.105.052008 (publication)
Subject category	hep-ex ; Particle Physics - Experiment ; cs.LG ; Computing and Computers ; cs.CV ; Computing and Computers ; physics.data-an ; Other Fields of Physics
Accelerator/Facility, Experiment	CERN LHC ; CMS
Abstract	We describe a novel application of the end-to-end deep learning technique to the task of discriminating top quark-initiated jets from those originating from the hadronization of a light quark or a gluon. The end-to-end deep learning technique combines deep learning algorithms and low-level detector representation of the high-energy collision event. In this study, we use low-level detector information from the simulated CMS Open Data samples to construct the top jet classifiers. To optimize classifier performance we progressively add low-level information from the CMS tracking detector, including pixel detector reconstructed hits and impact parameters, and demonstrate the value of additional tracking information even when no new spatial structures are added. Relying only on calorimeter energy deposits and reconstructed pixel detector hits, the end-to-end classifier achieves an AUC score of 0.975$\pm$0.002 for the task of classifying boosted top quark jets. After adding derived track quantities, the classifier AUC score increases to 0.9824$\pm$0.0013, serving as the first performance benchmark for these CMS Open Data samples. We additionally provide a timing performance comparison of different processor unit architectures for training the network.
Copyright/License	preprint: (License: CC BY 4.0) publication: © 2022-2024 American Physical Society (License: CC-BY-4.0)

Corresponding record in: Inspire