Cargando…

Deep Learning strategies for ProtoDUNE raw data denoising

In this work, we investigate different machine learning-based strategies for denoising raw simulation data from the ProtoDUNE experiment. The ProtoDUNE detector is hosted by CERN and it aims to test and calibrate the technologies for DUNE, a forthcoming experiment in neutrino physics. The reconstruc...

Descripción completa

Detalles Bibliográficos
Autores principales: Rossi, Marco, Vallecorsa, Sofia
Lenguaje:eng
Publicado: 2021
Materias:
Acceso en línea:https://dx.doi.org/10.1007/s41781-021-00077-9
http://cds.cern.ch/record/2758227
_version_ 1780970111508676608
author Rossi, Marco
Vallecorsa, Sofia
author_facet Rossi, Marco
Vallecorsa, Sofia
author_sort Rossi, Marco
collection CERN
description In this work, we investigate different machine learning-based strategies for denoising raw simulation data from the ProtoDUNE experiment. The ProtoDUNE detector is hosted by CERN and it aims to test and calibrate the technologies for DUNE, a forthcoming experiment in neutrino physics. The reconstruction workchain consists of converting digital detector signals into physical high-level quantities. We address the first step in reconstruction, namely raw data denoising, leveraging deep learning algorithms. We design two architectures based on graph neural networks, aiming to enhance the receptive field of basic convolutional neural networks. We benchmark this approach against traditional algorithms implemented by the DUNE collaboration. We test the capabilities of graph neural network hardware accelerator setups to speed up training and inference processes.
id cern-2758227
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2021
record_format invenio
spelling cern-27582272023-09-27T08:05:07Zdoi:10.1007/s41781-021-00077-9http://cds.cern.ch/record/2758227engRossi, MarcoVallecorsa, SofiaDeep Learning strategies for ProtoDUNE raw data denoisingstat.MLMathematical Physics and Mathematicsphysics.comp-phOther Fields of Physicshep-exParticle Physics - Experimentcs.LGComputing and Computershep-phParticle Physics - PhenomenologyIn this work, we investigate different machine learning-based strategies for denoising raw simulation data from the ProtoDUNE experiment. The ProtoDUNE detector is hosted by CERN and it aims to test and calibrate the technologies for DUNE, a forthcoming experiment in neutrino physics. The reconstruction workchain consists of converting digital detector signals into physical high-level quantities. We address the first step in reconstruction, namely raw data denoising, leveraging deep learning algorithms. We design two architectures based on graph neural networks, aiming to enhance the receptive field of basic convolutional neural networks. We benchmark this approach against traditional algorithms implemented by the DUNE collaboration. We test the capabilities of graph neural network hardware accelerator setups to speed up training and inference processes.In this work, we investigate different machine learning-based strategies for denoising raw simulation data from the ProtoDUNE experiment. The ProtoDUNE detector is hosted by CERN and it aims to test and calibrate the technologies for DUNE, a forthcoming experiment in neutrino physics. The reconstruction workchain consists of converting digital detector signals into physical high-level quantities. We address the first step in reconstruction, namely raw data denoising, leveraging deep learning algorithms. We design two architectures based on graph neural networks, aiming to enhance the receptive field of basic convolutional neural networks. We benchmark this approach against traditional algorithms implemented by the DUNE collaboration. We test the capabilities of graph neural network hardware accelerator setups to speed up training and inference processes.arXiv:2103.01596oai:cds.cern.ch:27582272021-03-02
spellingShingle stat.ML
Mathematical Physics and Mathematics
physics.comp-ph
Other Fields of Physics
hep-ex
Particle Physics - Experiment
cs.LG
Computing and Computers
hep-ph
Particle Physics - Phenomenology
Rossi, Marco
Vallecorsa, Sofia
Deep Learning strategies for ProtoDUNE raw data denoising
title Deep Learning strategies for ProtoDUNE raw data denoising
title_full Deep Learning strategies for ProtoDUNE raw data denoising
title_fullStr Deep Learning strategies for ProtoDUNE raw data denoising
title_full_unstemmed Deep Learning strategies for ProtoDUNE raw data denoising
title_short Deep Learning strategies for ProtoDUNE raw data denoising
title_sort deep learning strategies for protodune raw data denoising
topic stat.ML
Mathematical Physics and Mathematics
physics.comp-ph
Other Fields of Physics
hep-ex
Particle Physics - Experiment
cs.LG
Computing and Computers
hep-ph
Particle Physics - Phenomenology
url https://dx.doi.org/10.1007/s41781-021-00077-9
http://cds.cern.ch/record/2758227
work_keys_str_mv AT rossimarco deeplearningstrategiesforprotodunerawdatadenoising
AT vallecorsasofia deeplearningstrategiesforprotodunerawdatadenoising