Particle-based fast jet simulation at the LHC with variational autoencoders
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | , , , , , , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1088/2632-2153/ac7c56 http://hdl.handle.net/11449/240570 |
Resumo: | We study how to use deep variational autoencoders (VAEs) for a fast simulation of jets of particles at the Large Hadron Collider. We represent jets as a list of constituents, characterized by their momenta. Starting from a simulation of the jet before detector effects, we train a deep VAE to return the corresponding list of constituents after detection. Doing so, we bypass both the time-consuming detector simulation and the collision reconstruction steps of a traditional processing chain, speeding up significantly the events generation workflow. Through model optimization and hyperparameter tuning, we achieve state-of-the-art precision on the jet four-momentum, while providing an accurate description of the constituents momenta, and an inference time comparable to that of a rule-based fast simulation. |
id |
UNSP_2575e9a4c16ed030ad1cea429d2f71b1 |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/240570 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
Particle-based fast jet simulation at the LHC with variational autoencodersgenerative modelsparticle physicssparse data simulationWe study how to use deep variational autoencoders (VAEs) for a fast simulation of jets of particles at the Large Hadron Collider. We represent jets as a list of constituents, characterized by their momenta. Starting from a simulation of the jet before detector effects, we train a deep VAE to return the corresponding list of constituents after detection. Doing so, we bypass both the time-consuming detector simulation and the collision reconstruction steps of a traditional processing chain, speeding up significantly the events generation workflow. Through model optimization and hyperparameter tuning, we achieve state-of-the-art precision on the jet four-momentum, while providing an accurate description of the constituents momenta, and an inference time comparable to that of a rule-based fast simulation.European Organization for Nuclear Research (CERN)Department of Informatics and Telecommunications National and Kapodistrian University of AthensUniversity of California San Diego, La JollaUniversidade Estadual Paulista, SPCalifornia Institute of TechnologyUniversidade Estadual Paulista, SPEuropean Organization for Nuclear Research (CERN)National and Kapodistrian University of AthensUniversity of California San DiegoUniversidade Estadual Paulista (UNESP)California Institute of TechnologyTouranakou, MaryChernyavskaya, NadezdaDuarte, JavierGunopulos, DimitriosKansal, RaghavOrzari, Breno [UNESP]Pierini, MaurizioTomei, Thiago [UNESP]Vlimant, Jean-Roch2023-03-01T20:23:15Z2023-03-01T20:23:15Z2022-09-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articlehttp://dx.doi.org/10.1088/2632-2153/ac7c56Machine Learning: Science and Technology, v. 3, n. 3, 2022.2632-2153http://hdl.handle.net/11449/24057010.1088/2632-2153/ac7c562-s2.0-85135112343Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengMachine Learning: Science and Technologyinfo:eu-repo/semantics/openAccess2023-03-01T20:23:15Zoai:repositorio.unesp.br:11449/240570Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462023-03-01T20:23:15Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
Particle-based fast jet simulation at the LHC with variational autoencoders |
title |
Particle-based fast jet simulation at the LHC with variational autoencoders |
spellingShingle |
Particle-based fast jet simulation at the LHC with variational autoencoders Touranakou, Mary generative models particle physics sparse data simulation |
title_short |
Particle-based fast jet simulation at the LHC with variational autoencoders |
title_full |
Particle-based fast jet simulation at the LHC with variational autoencoders |
title_fullStr |
Particle-based fast jet simulation at the LHC with variational autoencoders |
title_full_unstemmed |
Particle-based fast jet simulation at the LHC with variational autoencoders |
title_sort |
Particle-based fast jet simulation at the LHC with variational autoencoders |
author |
Touranakou, Mary |
author_facet |
Touranakou, Mary Chernyavskaya, Nadezda Duarte, Javier Gunopulos, Dimitrios Kansal, Raghav Orzari, Breno [UNESP] Pierini, Maurizio Tomei, Thiago [UNESP] Vlimant, Jean-Roch |
author_role |
author |
author2 |
Chernyavskaya, Nadezda Duarte, Javier Gunopulos, Dimitrios Kansal, Raghav Orzari, Breno [UNESP] Pierini, Maurizio Tomei, Thiago [UNESP] Vlimant, Jean-Roch |
author2_role |
author author author author author author author author |
dc.contributor.none.fl_str_mv |
European Organization for Nuclear Research (CERN) National and Kapodistrian University of Athens University of California San Diego Universidade Estadual Paulista (UNESP) California Institute of Technology |
dc.contributor.author.fl_str_mv |
Touranakou, Mary Chernyavskaya, Nadezda Duarte, Javier Gunopulos, Dimitrios Kansal, Raghav Orzari, Breno [UNESP] Pierini, Maurizio Tomei, Thiago [UNESP] Vlimant, Jean-Roch |
dc.subject.por.fl_str_mv |
generative models particle physics sparse data simulation |
topic |
generative models particle physics sparse data simulation |
description |
We study how to use deep variational autoencoders (VAEs) for a fast simulation of jets of particles at the Large Hadron Collider. We represent jets as a list of constituents, characterized by their momenta. Starting from a simulation of the jet before detector effects, we train a deep VAE to return the corresponding list of constituents after detection. Doing so, we bypass both the time-consuming detector simulation and the collision reconstruction steps of a traditional processing chain, speeding up significantly the events generation workflow. Through model optimization and hyperparameter tuning, we achieve state-of-the-art precision on the jet four-momentum, while providing an accurate description of the constituents momenta, and an inference time comparable to that of a rule-based fast simulation. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-09-01 2023-03-01T20:23:15Z 2023-03-01T20:23:15Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.1088/2632-2153/ac7c56 Machine Learning: Science and Technology, v. 3, n. 3, 2022. 2632-2153 http://hdl.handle.net/11449/240570 10.1088/2632-2153/ac7c56 2-s2.0-85135112343 |
url |
http://dx.doi.org/10.1088/2632-2153/ac7c56 http://hdl.handle.net/11449/240570 |
identifier_str_mv |
Machine Learning: Science and Technology, v. 3, n. 3, 2022. 2632-2153 10.1088/2632-2153/ac7c56 2-s2.0-85135112343 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Machine Learning: Science and Technology |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.source.none.fl_str_mv |
Scopus reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1803045936433201152 |