Cargando…

LHC analysis-specific datasets with Generative Adversarial Networks

Using generative adversarial networks (GANs), we investigate the possibility of creating large amounts of analysis-specific simulated LHC events at limited computing cost. This kind of generative model is analysis specific in the sense that it directly generates the high-level features used in the l...

Descripción completa

Detalles Bibliográficos
Autores principales: Hashemi, Bobak, Amin, Nick, Datta, Kaustuv, Olivito, Dominick, Pierini, Maurizio
Lenguaje:eng
Publicado: 2019
Materias:
Acceso en línea:http://cds.cern.ch/record/2655426
_version_ 1780961171940048896
author Hashemi, Bobak
Amin, Nick
Datta, Kaustuv
Olivito, Dominick
Pierini, Maurizio
author_facet Hashemi, Bobak
Amin, Nick
Datta, Kaustuv
Olivito, Dominick
Pierini, Maurizio
author_sort Hashemi, Bobak
collection CERN
description Using generative adversarial networks (GANs), we investigate the possibility of creating large amounts of analysis-specific simulated LHC events at limited computing cost. This kind of generative model is analysis specific in the sense that it directly generates the high-level features used in the last stage of a given physics analyses, learning the N-dimensional distribution of relevant features in the context of a specific analysis selection. We apply this idea to the generation of muon four-momenta in $Z \to \mu\mu$ events at the LHC. We highlight how use-case specific issues emerge when the distributions of the considered quantities exhibit particular features. We show how substantial performance improvements and convergence speed-up can be obtained by including regression terms in the loss function of the generator. We develop an objective criterion to assess the geenrator performance in a quantitative way. With further development, a generalization of this approach could substantially reduce the needed amount of centrally produced fully simulated events in large particle physics experiments.
id cern-2655426
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2019
record_format invenio
spelling cern-26554262021-07-15T18:16:31Zhttp://cds.cern.ch/record/2655426engHashemi, BobakAmin, NickDatta, KaustuvOlivito, DominickPierini, MaurizioLHC analysis-specific datasets with Generative Adversarial Networkshep-phParticle Physics - Phenomenologycs.LGComputing and Computershep-exParticle Physics - ExperimentUsing generative adversarial networks (GANs), we investigate the possibility of creating large amounts of analysis-specific simulated LHC events at limited computing cost. This kind of generative model is analysis specific in the sense that it directly generates the high-level features used in the last stage of a given physics analyses, learning the N-dimensional distribution of relevant features in the context of a specific analysis selection. We apply this idea to the generation of muon four-momenta in $Z \to \mu\mu$ events at the LHC. We highlight how use-case specific issues emerge when the distributions of the considered quantities exhibit particular features. We show how substantial performance improvements and convergence speed-up can be obtained by including regression terms in the loss function of the generator. We develop an objective criterion to assess the geenrator performance in a quantitative way. With further development, a generalization of this approach could substantially reduce the needed amount of centrally produced fully simulated events in large particle physics experiments.arXiv:1901.05282oai:cds.cern.ch:26554262019
spellingShingle hep-ph
Particle Physics - Phenomenology
cs.LG
Computing and Computers
hep-ex
Particle Physics - Experiment
Hashemi, Bobak
Amin, Nick
Datta, Kaustuv
Olivito, Dominick
Pierini, Maurizio
LHC analysis-specific datasets with Generative Adversarial Networks
title LHC analysis-specific datasets with Generative Adversarial Networks
title_full LHC analysis-specific datasets with Generative Adversarial Networks
title_fullStr LHC analysis-specific datasets with Generative Adversarial Networks
title_full_unstemmed LHC analysis-specific datasets with Generative Adversarial Networks
title_short LHC analysis-specific datasets with Generative Adversarial Networks
title_sort lhc analysis-specific datasets with generative adversarial networks
topic hep-ph
Particle Physics - Phenomenology
cs.LG
Computing and Computers
hep-ex
Particle Physics - Experiment
url http://cds.cern.ch/record/2655426
work_keys_str_mv AT hashemibobak lhcanalysisspecificdatasetswithgenerativeadversarialnetworks
AT aminnick lhcanalysisspecificdatasetswithgenerativeadversarialnetworks
AT dattakaustuv lhcanalysisspecificdatasetswithgenerativeadversarialnetworks
AT olivitodominick lhcanalysisspecificdatasetswithgenerativeadversarialnetworks
AT pierinimaurizio lhcanalysisspecificdatasetswithgenerativeadversarialnetworks