Cargando…
Sparse Data Generation for Particle-Based Simulation of Hadronic Jets in the LHC
We develop a generative neural network for the generation of sparse data in particle physics using a permutation-invariant and physics-informed loss function. The input dataset used in this study consists of the particle constituents of hadronic jets due to its sparsity and the possibility of evalua...
Autores principales: | , , , , , , , |
---|---|
Lenguaje: | eng |
Publicado: |
2021
|
Materias: | |
Acceso en línea: | http://cds.cern.ch/record/2784343 |
_version_ | 1780972097734967296 |
---|---|
author | Orzari, Breno Tomei, Thiago Pierini, Maurizio Touranakou, Mary Duarte, Javier Kansal, Raghav Vlimant, Jean-Roch Gunopulos, Dimitrios |
author_facet | Orzari, Breno Tomei, Thiago Pierini, Maurizio Touranakou, Mary Duarte, Javier Kansal, Raghav Vlimant, Jean-Roch Gunopulos, Dimitrios |
author_sort | Orzari, Breno |
collection | CERN |
description | We develop a generative neural network for the generation of sparse data in particle physics using a permutation-invariant and physics-informed loss function. The input dataset used in this study consists of the particle constituents of hadronic jets due to its sparsity and the possibility of evaluating the network's ability to accurately describe the particles and jets properties. A variational autoencoder composed of convolutional layers in the encoder and decoder is used as the generator. The loss function consists of a reconstruction error term and the Kullback-Leibler divergence between the output of the encoder and the latent vector variables. The permutation-invariant loss on the particles' properties is combined with two mean-squared error terms that measure the difference between input and output jets mass and transverse momentum, which improves the network's generation capability as it imposes physics constraints, allowing the model to learn the kinematics of the jets. |
id | cern-2784343 |
institution | Organización Europea para la Investigación Nuclear |
language | eng |
publishDate | 2021 |
record_format | invenio |
spelling | cern-27843432023-01-31T10:25:30Zhttp://cds.cern.ch/record/2784343engOrzari, BrenoTomei, ThiagoPierini, MaurizioTouranakou, MaryDuarte, JavierKansal, RaghavVlimant, Jean-RochGunopulos, DimitriosSparse Data Generation for Particle-Based Simulation of Hadronic Jets in the LHChep-exParticle Physics - Experimentphysics.data-anOther Fields of PhysicsWe develop a generative neural network for the generation of sparse data in particle physics using a permutation-invariant and physics-informed loss function. The input dataset used in this study consists of the particle constituents of hadronic jets due to its sparsity and the possibility of evaluating the network's ability to accurately describe the particles and jets properties. A variational autoencoder composed of convolutional layers in the encoder and decoder is used as the generator. The loss function consists of a reconstruction error term and the Kullback-Leibler divergence between the output of the encoder and the latent vector variables. The permutation-invariant loss on the particles' properties is combined with two mean-squared error terms that measure the difference between input and output jets mass and transverse momentum, which improves the network's generation capability as it imposes physics constraints, allowing the model to learn the kinematics of the jets.arXiv:2109.15197oai:cds.cern.ch:27843432021-09-30 |
spellingShingle | hep-ex Particle Physics - Experiment physics.data-an Other Fields of Physics Orzari, Breno Tomei, Thiago Pierini, Maurizio Touranakou, Mary Duarte, Javier Kansal, Raghav Vlimant, Jean-Roch Gunopulos, Dimitrios Sparse Data Generation for Particle-Based Simulation of Hadronic Jets in the LHC |
title | Sparse Data Generation for Particle-Based Simulation of Hadronic Jets in the LHC |
title_full | Sparse Data Generation for Particle-Based Simulation of Hadronic Jets in the LHC |
title_fullStr | Sparse Data Generation for Particle-Based Simulation of Hadronic Jets in the LHC |
title_full_unstemmed | Sparse Data Generation for Particle-Based Simulation of Hadronic Jets in the LHC |
title_short | Sparse Data Generation for Particle-Based Simulation of Hadronic Jets in the LHC |
title_sort | sparse data generation for particle-based simulation of hadronic jets in the lhc |
topic | hep-ex Particle Physics - Experiment physics.data-an Other Fields of Physics |
url | http://cds.cern.ch/record/2784343 |
work_keys_str_mv | AT orzaribreno sparsedatagenerationforparticlebasedsimulationofhadronicjetsinthelhc AT tomeithiago sparsedatagenerationforparticlebasedsimulationofhadronicjetsinthelhc AT pierinimaurizio sparsedatagenerationforparticlebasedsimulationofhadronicjetsinthelhc AT touranakoumary sparsedatagenerationforparticlebasedsimulationofhadronicjetsinthelhc AT duartejavier sparsedatagenerationforparticlebasedsimulationofhadronicjetsinthelhc AT kansalraghav sparsedatagenerationforparticlebasedsimulationofhadronicjetsinthelhc AT vlimantjeanroch sparsedatagenerationforparticlebasedsimulationofhadronicjetsinthelhc AT gunopulosdimitrios sparsedatagenerationforparticlebasedsimulationofhadronicjetsinthelhc |