Cargando…

Machine Learning for Particle Flow Reconstruction at CMS

We provide details on the implementation of a machine-learning based particle flow algorithm for CMS. The standard particle flow algorithm reconstructs stable particles based on calorimeter clusters and tracks to provide a global event reconstruction that exploits the combined information of multipl...

Descripción completa

Detalles Bibliográficos
Autores principales: Pata, Joosep, Duarte, Javier, Mokhtar, Farouk, Wulff, Eric, Yoo, Jieun, Vlimant, Jean-Roch, Pierini, Maurizio, Girone, Maria
Lenguaje:eng
Publicado: 2023
Materias:
Acceso en línea:https://dx.doi.org/10.1088/1742-6596/2438/1/012100
http://cds.cern.ch/record/2802826
_version_ 1780972761213042688
author Pata, Joosep
Duarte, Javier
Mokhtar, Farouk
Wulff, Eric
Yoo, Jieun
Vlimant, Jean-Roch
Pierini, Maurizio
Girone, Maria
author_facet Pata, Joosep
Duarte, Javier
Mokhtar, Farouk
Wulff, Eric
Yoo, Jieun
Vlimant, Jean-Roch
Pierini, Maurizio
Girone, Maria
author_sort Pata, Joosep
collection CERN
description We provide details on the implementation of a machine-learning based particle flow algorithm for CMS. The standard particle flow algorithm reconstructs stable particles based on calorimeter clusters and tracks to provide a global event reconstruction that exploits the combined information of multiple detector subsystems, leading to strong improvements for quantities such as jets and missing transverse energy. We have studied a possible evolution of particle flow towards heterogeneous computing platforms such as GPUs using a graph neural network. The machine-learned PF model reconstructs particle candidates based on the full list of tracks and calorimeter clusters in the event. For validation, we determine the physics performance directly in the CMS software framework when the proposed algorithm is interfaced with the offline reconstruction of jets and missing transverse energy. We also report the computational performance of the algorithm, which scales approximately linearly in runtime and memory usage with the input size.
id cern-2802826
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2023
record_format invenio
spelling cern-28028262023-09-27T07:59:06Zdoi:10.1088/1742-6596/2438/1/012100http://cds.cern.ch/record/2802826engPata, JoosepDuarte, JavierMokhtar, FaroukWulff, EricYoo, JieunVlimant, Jean-RochPierini, MaurizioGirone, MariaMachine Learning for Particle Flow Reconstruction at CMSMathematical Physics and MathematicsDetectors and Experimental TechniquesParticle Physics - ExperimentComputing and ComputersOther Fields of PhysicsWe provide details on the implementation of a machine-learning based particle flow algorithm for CMS. The standard particle flow algorithm reconstructs stable particles based on calorimeter clusters and tracks to provide a global event reconstruction that exploits the combined information of multiple detector subsystems, leading to strong improvements for quantities such as jets and missing transverse energy. We have studied a possible evolution of particle flow towards heterogeneous computing platforms such as GPUs using a graph neural network. The machine-learned PF model reconstructs particle candidates based on the full list of tracks and calorimeter clusters in the event. For validation, we determine the physics performance directly in the CMS software framework when the proposed algorithm is interfaced with the offline reconstruction of jets and missing transverse energy. We also report the computational performance of the algorithm, which scales approximately linearly in runtime and memory usage with the input size.We provide details on the implementation of a machine-learning based particle flow algorithm for CMS. The standard particle flow algorithm reconstructs stable particles based on calorimeter clusters and tracks to provide a global event reconstruction that exploits the combined information of multiple detector subsystems, leading to strong improvements for quantities such as jets and missing transverse energy. We have studied a possible evolution of particle flow towards heterogeneous computing platforms such as GPUs using a graph neural network. The machine-learned PF model reconstructs particle candidates based on the full list of tracks and calorimeter clusters in the event. For validation, we determine the physics performance directly in the CMS software framework when the proposed algorithm is interfaced with the offline reconstruction of jets and missing transverse energy. We also report the computational performance of the algorithm, which scales approximately linearly in runtime and memory usage with the input size.arXiv:2203.00330CMS-CR-2022-021oai:cds.cern.ch:28028262023
spellingShingle Mathematical Physics and Mathematics
Detectors and Experimental Techniques
Particle Physics - Experiment
Computing and Computers
Other Fields of Physics
Pata, Joosep
Duarte, Javier
Mokhtar, Farouk
Wulff, Eric
Yoo, Jieun
Vlimant, Jean-Roch
Pierini, Maurizio
Girone, Maria
Machine Learning for Particle Flow Reconstruction at CMS
title Machine Learning for Particle Flow Reconstruction at CMS
title_full Machine Learning for Particle Flow Reconstruction at CMS
title_fullStr Machine Learning for Particle Flow Reconstruction at CMS
title_full_unstemmed Machine Learning for Particle Flow Reconstruction at CMS
title_short Machine Learning for Particle Flow Reconstruction at CMS
title_sort machine learning for particle flow reconstruction at cms
topic Mathematical Physics and Mathematics
Detectors and Experimental Techniques
Particle Physics - Experiment
Computing and Computers
Other Fields of Physics
url https://dx.doi.org/10.1088/1742-6596/2438/1/012100
http://cds.cern.ch/record/2802826
work_keys_str_mv AT patajoosep machinelearningforparticleflowreconstructionatcms
AT duartejavier machinelearningforparticleflowreconstructionatcms
AT mokhtarfarouk machinelearningforparticleflowreconstructionatcms
AT wulfferic machinelearningforparticleflowreconstructionatcms
AT yoojieun machinelearningforparticleflowreconstructionatcms
AT vlimantjeanroch machinelearningforparticleflowreconstructionatcms
AT pierinimaurizio machinelearningforparticleflowreconstructionatcms
AT gironemaria machinelearningforparticleflowreconstructionatcms