Deep-learned Top Tagging with a Lorentz Layer

We introduce a new and highly efficient tagger for hadronically decaying top quarks, based on a deep neural network working with Lorentz vectors and the Minkowski metric. With its novel machine learning setup and architecture it allows us to identify boosted top quarks not only from calorimeter t...

Full description

Bibliographic Details
Main Author:	Anja Butter, Gregor Kasieczka, Tilman Plehn, Michael Russell
Format:	Article
Language:	English
Published:	SciPost 2018-09-01
Series:	SciPost Physics
Online Access:	https://scipost.org/SciPostPhys.5.3.028

id	doaj-011702377e07410f9cfe48db916a295a
record_format	Article
spelling	doaj-011702377e07410f9cfe48db916a295a2020-11-24T22:59:19ZengSciPostSciPost Physics2542-46532018-09-015302810.21468/SciPostPhys.5.3.028Deep-learned Top Tagging with a Lorentz LayerAnja Butter, Gregor Kasieczka, Tilman Plehn, Michael RussellWe introduce a new and highly efficient tagger for hadronically decaying top quarks, based on a deep neural network working with Lorentz vectors and the Minkowski metric. With its novel machine learning setup and architecture it allows us to identify boosted top quarks not only from calorimeter towers, but also including tracking information. We show how the performance of our tagger compares with QCD-inspired and image-recognition approaches and find that it significantly increases the performance for strongly boosted top quarks.https://scipost.org/SciPostPhys.5.3.028
collection	DOAJ
language	English
format	Article
sources	DOAJ
author	Anja Butter, Gregor Kasieczka, Tilman Plehn, Michael Russell
spellingShingle	Anja Butter, Gregor Kasieczka, Tilman Plehn, Michael Russell Deep-learned Top Tagging with a Lorentz Layer SciPost Physics
author_facet	Anja Butter, Gregor Kasieczka, Tilman Plehn, Michael Russell
author_sort	Anja Butter, Gregor Kasieczka, Tilman Plehn, Michael Russell
title	Deep-learned Top Tagging with a Lorentz Layer
title_short	Deep-learned Top Tagging with a Lorentz Layer
title_full	Deep-learned Top Tagging with a Lorentz Layer
title_fullStr	Deep-learned Top Tagging with a Lorentz Layer
title_full_unstemmed	Deep-learned Top Tagging with a Lorentz Layer
title_sort	deep-learned top tagging with a lorentz layer
publisher	SciPost
series	SciPost Physics
issn	2542-4653
publishDate	2018-09-01
description	We introduce a new and highly efficient tagger for hadronically decaying top quarks, based on a deep neural network working with Lorentz vectors and the Minkowski metric. With its novel machine learning setup and architecture it allows us to identify boosted top quarks not only from calorimeter towers, but also including tracking information. We show how the performance of our tagger compares with QCD-inspired and image-recognition approaches and find that it significantly increases the performance for strongly boosted top quarks.
url	https://scipost.org/SciPostPhys.5.3.028
work_keys_str_mv	AT anjabuttergregorkasieczkatilmanplehnmichaelrussell deeplearnedtoptaggingwithalorentzlayer
_version_	1725645022887411712

Deep-learned Top Tagging with a Lorentz Layer

Similar Items