Cargando…
Optimizing electron and photon reconstruction using deep learning application to the CMS electromagnetic calorimeter
The reconstruction of electrons and photons in CMS depends on topological clustering of the energy deposited by an incident particle in different crystals of the electromagnetic calorimeter (ECAL). These clusters are formed by aggregating neighbouring crystals according to the expected topology of a...
Autor principal: | |
---|---|
Lenguaje: | eng |
Publicado: |
2023
|
Materias: | |
Acceso en línea: | http://cds.cern.ch/record/2858893 |
_version_ | 1780977678922285056 |
---|---|
author | Valsecchi, Davide |
author_facet | Valsecchi, Davide |
author_sort | Valsecchi, Davide |
collection | CERN |
description | The reconstruction of electrons and photons in CMS depends on
topological clustering of the energy deposited by an incident particle
in different crystals of the electromagnetic calorimeter (ECAL). These
clusters are formed by aggregating neighbouring crystals according to
the expected topology of an electromagnetic shower in the ECAL. The
presence of upstream material (beampipe, tracker, and support structures)
causes electrons and photons to start showering before reaching the
calorimeter. This effect, combined with the 3.8\,T CMS magnetic field,
leads to energy being spread in several clusters around the primary one.
It is essential to recover the energy contained in these satellite
clusters in order to achieve the best possible energy resolution for
physics analyses.
Historically, satellite clusters have been associated to the primary
cluster using a purely topological algorithm which does not attempt to
remove spurious energy deposits from additional pileup interactions
(PU). The performance of this algorithm is expected to degrade during
LHC Run 3 (started in 2022) because of the larger average PU and the
increasing levels of noise due to the ageing of the ECAL detector. New
methods are being investigated, which exploit state-of-the-art deep
learning architectures like Graph Neural Networks (GNN). These
more sophisticated models improve the energy collection and are
more resilient to PU and noise.
This contribution covers the results of the model optimization,
and the stpes to deploy it in the CMS reconstruction sequence.
The inference performance is analyzed and strategies
to improve it are described in this manuscript. |
id | cern-2858893 |
institution | Organización Europea para la Investigación Nuclear |
language | eng |
publishDate | 2023 |
record_format | invenio |
spelling | cern-28588932023-05-15T19:37:35Zhttp://cds.cern.ch/record/2858893engValsecchi, DavideOptimizing electron and photon reconstruction using deep learning application to the CMS electromagnetic calorimeterDetectors and Experimental TechniquesThe reconstruction of electrons and photons in CMS depends on topological clustering of the energy deposited by an incident particle in different crystals of the electromagnetic calorimeter (ECAL). These clusters are formed by aggregating neighbouring crystals according to the expected topology of an electromagnetic shower in the ECAL. The presence of upstream material (beampipe, tracker, and support structures) causes electrons and photons to start showering before reaching the calorimeter. This effect, combined with the 3.8\,T CMS magnetic field, leads to energy being spread in several clusters around the primary one. It is essential to recover the energy contained in these satellite clusters in order to achieve the best possible energy resolution for physics analyses. Historically, satellite clusters have been associated to the primary cluster using a purely topological algorithm which does not attempt to remove spurious energy deposits from additional pileup interactions (PU). The performance of this algorithm is expected to degrade during LHC Run 3 (started in 2022) because of the larger average PU and the increasing levels of noise due to the ageing of the ECAL detector. New methods are being investigated, which exploit state-of-the-art deep learning architectures like Graph Neural Networks (GNN). These more sophisticated models improve the energy collection and are more resilient to PU and noise. This contribution covers the results of the model optimization, and the stpes to deploy it in the CMS reconstruction sequence. The inference performance is analyzed and strategies to improve it are described in this manuscript.CMS-CR-2023-039oai:cds.cern.ch:28588932023-03-16 |
spellingShingle | Detectors and Experimental Techniques Valsecchi, Davide Optimizing electron and photon reconstruction using deep learning application to the CMS electromagnetic calorimeter |
title | Optimizing electron and photon reconstruction using deep learning application to the CMS electromagnetic calorimeter |
title_full | Optimizing electron and photon reconstruction using deep learning application to the CMS electromagnetic calorimeter |
title_fullStr | Optimizing electron and photon reconstruction using deep learning application to the CMS electromagnetic calorimeter |
title_full_unstemmed | Optimizing electron and photon reconstruction using deep learning application to the CMS electromagnetic calorimeter |
title_short | Optimizing electron and photon reconstruction using deep learning application to the CMS electromagnetic calorimeter |
title_sort | optimizing electron and photon reconstruction using deep learning application to the cms electromagnetic calorimeter |
topic | Detectors and Experimental Techniques |
url | http://cds.cern.ch/record/2858893 |
work_keys_str_mv | AT valsecchidavide optimizingelectronandphotonreconstructionusingdeeplearningapplicationtothecmselectromagneticcalorimeter |