% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@TECHREPORT{Collaboration:600611,
key = {600611},
collaboration = {{ATLAS Collaboration}},
title = {{E}lectron {I}dentification with a {C}onvolutional {N}eural
{N}etwork in the {ATLAS} {E}xperiment},
number = {ATL-PHYS-PUB-2023-001},
reportid = {PUBDB-2023-08035, ATL-PHYS-PUB-2023-001},
pages = {33},
year = {2023},
note = {All 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!},
abstract = {The 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.},
keywords = {p p: scattering (INSPIRE) / p p: colliding beams (INSPIRE)
/ electron: background (INSPIRE) / electron: particle
identification (INSPIRE) / hadron: decay (INSPIRE) /
calorimeter: hadronic (INSPIRE) / ATLAS (INSPIRE) / neural
network (INSPIRE) / network (INSPIRE) / calorimeter:
electromagnetic (INSPIRE) / tracks (INSPIRE) / machine
learning (INSPIRE) / particle identification: efficiency
(INSPIRE) / CERN LHC Coll (INSPIRE) / kinematics (INSPIRE) /
particle identification: performance (INSPIRE) / heavy
quark: decay (INSPIRE) / statistical analysis (INSPIRE) /
data analysis method (INSPIRE) / numerical calculations:
Monte Carlo (INSPIRE) / experimental results (INSPIRE) /
EGAMMA (autogen)},
cin = {ATLAS},
cid = {I:(DE-H253)ATLAS-20120731},
pnm = {611 - Fundamental Particles and Forces (POF4-611)},
pid = {G:(DE-HGF)POF4-611},
experiment = {EXP:(DE-H253)LHC-Exp-ATLAS-20150101},
typ = {PUB:(DE-HGF)29},
doi = {10.3204/PUBDB-2023-08035},
url = {https://bib-pubdb1.desy.de/record/600611},
}