Deep-learning top taggers or the end of QCD?

• Verfasst von: Kasieczka, Gregor [VerfasserIn]
• Verfasst von: Plehn, Tilman [VerfasserIn]
• Verfasst von: Schell, Torben [VerfasserIn]
• Titel: Deep-learning top taggers or the end of QCD?
• Verf.angabe: Gregor Kasieczka, Tilman Plehn, Michael Russell and Torben Schell
• Fussnoten: Gesehen am 26.09.2017
• Titel Quelle: Enthalten in: Journal of high energy physics
• Jahr Quelle: 2017
• Band/Heft Quelle: (2017,05) Artikel-Nummer 006, 22 Seiten
• ISSN Quelle: 1029-8479
• DOI: doi:10.1007/JHEP05(2017)006
• Datenträger: Online-Ressource
• Sprache: eng
• K10plus-PPN: 1562316796

Abstract

Machine learning based on convolutional neural networks can be used to study jet images from the LHC. Top tagging in fat jets offers a well-defined framework to establish our DeepTop approach and compare its performance to QCD-based top taggers. We first optimize a network architecture to identify top quarks in Monte Carlo simulations of the Standard Model production channel. Using standard fat jets we then compare its performance to a multivariate QCD-based top tagger. We find that both approaches lead to comparable performance, establishing convolutional networks as a promising new approach for multivariate hypothesis-based top tagging.