Performance of top-quark and W -boson tagging with ATLAS in Run 2 of the LHC

ATLAS Collaboration

doi:10.1140/epjc/s10052-019-6847-8

Performance of top-quark and W -boson tagging with ATLAS in Run 2 of the LHC

ATLAS Collaboration

Research output: Contribution to journal › Article › peer-review

80 Citations (Scopus)

Abstract

The performance of identification algorithms (“taggers”) for hadronically decaying top quarks and W bosons in pp collisions at s = 13 TeV recorded by the ATLAS experiment at the Large Hadron Collider is presented. A set of techniques based on jet shape observables are studied to determine a set of optimal cut-based taggers for use in physics analyses. The studies are extended to assess the utility of combinations of substructure observables as a multivariate tagger using boosted decision trees or deep neural networks in comparison with taggers based on two-variable combinations. In addition, for highly boosted top-quark tagging, a deep neural network based on jet constituent inputs as well as a re-optimisation of the shower deconstruction technique is presented. The performance of these taggers is studied in data collected during 2015 and 2016 corresponding to 36.1 fb ^{- 1} for the tt¯ and γ+ jet and 36.7 fb ^{- 1} for the dijet event topologies.

Original language	English
Article number	375
Journal	European Physical Journal C
Volume	79
Issue number	5
DOIs	https://doi.org/10.1140/epjc/s10052-019-6847-8
Publication status	Published - 1 May 2019

ASJC Scopus subject areas

Engineering (miscellaneous)
Physics and Astronomy (miscellaneous)

Access to Document

10.1140/epjc/s10052-019-6847-8

Cite this

@article{ca213e3167354891b5cd3ae395579686,

title = "Performance of top-quark and W -boson tagging with ATLAS in Run 2 of the LHC",

abstract = " The performance of identification algorithms (“taggers”) for hadronically decaying top quarks and W bosons in pp collisions at s = 13 TeV recorded by the ATLAS experiment at the Large Hadron Collider is presented. A set of techniques based on jet shape observables are studied to determine a set of optimal cut-based taggers for use in physics analyses. The studies are extended to assess the utility of combinations of substructure observables as a multivariate tagger using boosted decision trees or deep neural networks in comparison with taggers based on two-variable combinations. In addition, for highly boosted top-quark tagging, a deep neural network based on jet constituent inputs as well as a re-optimisation of the shower deconstruction technique is presented. The performance of these taggers is studied in data collected during 2015 and 2016 corresponding to 36.1 fb - 1 for the tt¯ and γ+ jet and 36.7 fb - 1 for the dijet event topologies.",

author = "{ATLAS Collaboration} and M. Aaboud and G. Aad and B. Abbott and O. Abdinov and B. Abeloos and Abhayasinghe, {D. K.} and Abidi, {S. H.} and AbouZeid, {O. S.} and Abraham, {N. L.} and H. Abramowicz and H. Abreu and Y. Abulaiti and Acharya, {B. S.} and S. Adachi and L. Adam and L. Adamczyk and J. Adelman and M. Adersberger and A. Adiguzel and T. Adye and Affolder, {A. A.} and Y. Afik and C. Agheorghiesei and Aguilar-Saavedra, {J. A.} and F. Ahmadov and G. Aielli and S. Akatsuka and {\AA}kesson, {T. P.A.} and E. Akilli and Akimov, {A. V.} and Alberghi, {G. L.} and J. Albert and P. Albicocco and Alconada Verzini, {M. J.} and S. Alderweireldt and M. Aleksa and Aleksandrov, {I. N.} and C. Alexa and T. Alexopoulos and M. Alhroob and B. Ali and G. Alimonti and J. Alison and Alkire, {S. P.} and C. Allaire and Allbrooke, {B. M.M.} and Allen, {B. W.} and Allport, {P. P.} and Connell, {S. H.} and N. Govender",

note = "Publisher Copyright: {\textcopyright} 2019, CERN for the benefit of the ATLAS collaboration.",

year = "2019",

month = may,

day = "1",

doi = "10.1140/epjc/s10052-019-6847-8",

language = "English",

volume = "79",

journal = "European Physical Journal C",

issn = "1434-6044",

publisher = "Springer New York",

number = "5",

}

TY - JOUR

T1 - Performance of top-quark and W -boson tagging with ATLAS in Run 2 of the LHC

AU - ATLAS Collaboration

AU - Aaboud, M.

AU - Aad, G.

AU - Abbott, B.

AU - Abdinov, O.

AU - Abeloos, B.

AU - Abhayasinghe, D. K.

AU - Abidi, S. H.

AU - AbouZeid, O. S.

AU - Abraham, N. L.

AU - Abramowicz, H.

AU - Abreu, H.

AU - Abulaiti, Y.

AU - Acharya, B. S.

AU - Adachi, S.

AU - Adam, L.

AU - Adamczyk, L.

AU - Adelman, J.

AU - Adersberger, M.

AU - Adiguzel, A.

AU - Adye, T.

AU - Affolder, A. A.

AU - Afik, Y.

AU - Agheorghiesei, C.

AU - Aguilar-Saavedra, J. A.

AU - Ahmadov, F.

AU - Aielli, G.

AU - Akatsuka, S.

AU - Åkesson, T. P.A.

AU - Akilli, E.

AU - Akimov, A. V.

AU - Alberghi, G. L.

AU - Albert, J.

AU - Albicocco, P.

AU - Alconada Verzini, M. J.

AU - Alderweireldt, S.

AU - Aleksa, M.

AU - Aleksandrov, I. N.

AU - Alexa, C.

AU - Alexopoulos, T.

AU - Alhroob, M.

AU - Ali, B.

AU - Alimonti, G.

AU - Alison, J.

AU - Alkire, S. P.

AU - Allaire, C.

AU - Allbrooke, B. M.M.

AU - Allen, B. W.

AU - Allport, P. P.

AU - Connell, S. H.

AU - Govender, N.

PY - 2019/5/1

Y1 - 2019/5/1

N2 - The performance of identification algorithms (“taggers”) for hadronically decaying top quarks and W bosons in pp collisions at s = 13 TeV recorded by the ATLAS experiment at the Large Hadron Collider is presented. A set of techniques based on jet shape observables are studied to determine a set of optimal cut-based taggers for use in physics analyses. The studies are extended to assess the utility of combinations of substructure observables as a multivariate tagger using boosted decision trees or deep neural networks in comparison with taggers based on two-variable combinations. In addition, for highly boosted top-quark tagging, a deep neural network based on jet constituent inputs as well as a re-optimisation of the shower deconstruction technique is presented. The performance of these taggers is studied in data collected during 2015 and 2016 corresponding to 36.1 fb - 1 for the tt¯ and γ+ jet and 36.7 fb - 1 for the dijet event topologies.

AB - The performance of identification algorithms (“taggers”) for hadronically decaying top quarks and W bosons in pp collisions at s = 13 TeV recorded by the ATLAS experiment at the Large Hadron Collider is presented. A set of techniques based on jet shape observables are studied to determine a set of optimal cut-based taggers for use in physics analyses. The studies are extended to assess the utility of combinations of substructure observables as a multivariate tagger using boosted decision trees or deep neural networks in comparison with taggers based on two-variable combinations. In addition, for highly boosted top-quark tagging, a deep neural network based on jet constituent inputs as well as a re-optimisation of the shower deconstruction technique is presented. The performance of these taggers is studied in data collected during 2015 and 2016 corresponding to 36.1 fb - 1 for the tt¯ and γ+ jet and 36.7 fb - 1 for the dijet event topologies.

UR - http://www.scopus.com/inward/record.url?scp=85065123030&partnerID=8YFLogxK

U2 - 10.1140/epjc/s10052-019-6847-8

DO - 10.1140/epjc/s10052-019-6847-8

M3 - Article

AN - SCOPUS:85065123030

SN - 1434-6044

VL - 79

JO - European Physical Journal C

JF - European Physical Journal C

IS - 5

M1 - 375

ER -

Performance of top-quark and W -boson tagging with ATLAS in Run 2 of the LHC

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this