000600611 001__ 600611
000600611 005__ 20231221211936.0
000600611 0247_ $$2INSPIRETeX$$aATLAS:2023mnn
000600611 0247_ $$2inspire$$ainspire:2636611
000600611 0247_ $$2datacite_doi$$a10.3204/PUBDB-2023-08035
000600611 037__ $$aPUBDB-2023-08035
000600611 041__ $$aEnglish
000600611 088__ $$2Atlas$$aATL-PHYS-PUB-2023-001
000600611 1001_ $$0P:(DE-HGF)0$$aATLAS Collaboration$$b0$$eCollaboration author
000600611 245__ $$aElectron Identification with a Convolutional Neural Network in the ATLAS Experiment
000600611 260__ $$c2023
000600611 300__ $$a33
000600611 3367_ $$2DRIVER$$areport
000600611 3367_ $$2ORCID$$aREPORT
000600611 3367_ $$010$$2EndNote$$aReport
000600611 3367_ $$2DataCite$$aOutput Types/Report
000600611 3367_ $$0PUB:(DE-HGF)29$$2PUB:(DE-HGF)$$aReport$$breport$$mreport$$s1703152023_12800
000600611 3367_ $$2BibTeX$$aTECHREPORT
000600611 500__ $$aAll figures including auxiliary figures are available at https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PUBNOTES/ATL-PHYS-PUB-2023-001. The authors list may be incomplete!
000600611 520__ $$aThe identification of electrons plays an important role for a large fraction of the physics analyses performed at the ATLAS experiment. An improved electron identification algorithm is presented that is based on convolutional neural networks (CNN), a type of machine learning architecture specialized in image recognition. It takes as input the images of the deposited energy in the calorimeter cells around the reconstructed electron candidates for each of the electromagnetic and hadronic calorimeter layers. Additional input features include the same high-level variables that are used by the likelihood (LLH) and deep neural network (DNN) algorithms developed in ATLAS, as well as the information of up to five inner detector tracks that are matched to an electron candidate during its reconstruction. The output of the network corresponds to the probability that a reconstructed electron belongs to six classes of signals and backgrounds. A significant improvement in identification performance is observed when the CNN algorithm is used in the simulation. For example, for a working point that corresponds to the same signal efficiency as the LLH 'Loose' working point, the CNN improves the rejection against charged hadrons faking signal electrons, the dominant electron background at the LHC, by factors of 5 to 8 (depending on the electron kinematics) with respect to the LLH. For the most difficult background constituted of electrons originating from heavy flavour hadron decays, the background rejection of the CNN is improved by factors varying between about 2 to 3.5 with respect to the LLH.
000600611 536__ $$0G:(DE-HGF)POF4-611$$a611 - Fundamental Particles and Forces (POF4-611)$$cPOF4-611$$fPOF IV$$x0
000600611 588__ $$aDataset connected to INSPIRE
000600611 650_7 $$2INSPIRE$$ap p: scattering
000600611 650_7 $$2INSPIRE$$ap p: colliding beams
000600611 650_7 $$2INSPIRE$$aelectron: background
000600611 650_7 $$2INSPIRE$$aelectron: particle identification
000600611 650_7 $$2INSPIRE$$ahadron: decay
000600611 650_7 $$2INSPIRE$$acalorimeter: hadronic
000600611 650_7 $$2INSPIRE$$aATLAS
000600611 650_7 $$2INSPIRE$$aneural network
000600611 650_7 $$2INSPIRE$$anetwork
000600611 650_7 $$2INSPIRE$$acalorimeter: electromagnetic
000600611 650_7 $$2INSPIRE$$atracks
000600611 650_7 $$2INSPIRE$$amachine learning
000600611 650_7 $$2INSPIRE$$aparticle identification: efficiency
000600611 650_7 $$2INSPIRE$$aCERN LHC Coll
000600611 650_7 $$2INSPIRE$$akinematics
000600611 650_7 $$2INSPIRE$$aparticle identification: performance
000600611 650_7 $$2INSPIRE$$aheavy quark: decay
000600611 650_7 $$2INSPIRE$$astatistical analysis
000600611 650_7 $$2INSPIRE$$adata analysis method
000600611 650_7 $$2INSPIRE$$anumerical calculations: Monte Carlo
000600611 650_7 $$2INSPIRE$$aexperimental results
000600611 650_7 $$2autogen$$aEGAMMA
000600611 693__ $$0EXP:(DE-H253)LHC-Exp-ATLAS-20150101$$1EXP:(DE-588)4398783-7$$5EXP:(DE-H253)LHC-Exp-ATLAS-20150101$$aLHC$$eLHC: ATLAS$$x0
000600611 8564_ $$uhttps://bib-pubdb1.desy.de/record/600611/files/ATL-PHYS-PUB-2023-001.pdf$$yOpenAccess
000600611 8564_ $$uhttps://bib-pubdb1.desy.de/record/600611/files/ATL-PHYS-PUB-2023-001.pdf?subformat=pdfa$$xpdfa$$yOpenAccess
000600611 909CO $$ooai:bib-pubdb1.desy.de:600611$$popenaire$$popen_access$$pVDB$$pdriver$$pdnbdelivery
000600611 9101_ $$0I:(DE-588b)2008985-5$$6P:(DE-HGF)0$$aDeutsches Elektronen-Synchrotron$$b0$$kDESY
000600611 9131_ $$0G:(DE-HGF)POF4-611$$1G:(DE-HGF)POF4-610$$2G:(DE-HGF)POF4-600$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$aDE-HGF$$bForschungsbereich Materie$$lMatter and the Universe$$vFundamental Particles and Forces$$x0
000600611 9141_ $$y2023
000600611 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
000600611 915__ $$0LIC:(DE-HGF)CCBY4$$2HGFVOC$$aCreative Commons Attribution CC BY 4.0
000600611 9201_ $$0I:(DE-H253)ATLAS-20120731$$kATLAS$$lLHC/ATLAS Experiment$$x0
000600611 980__ $$areport
000600611 980__ $$aVDB
000600611 980__ $$aUNRESTRICTED
000600611 980__ $$aI:(DE-H253)ATLAS-20120731
000600611 9801_ $$aFullTexts