Cargando…

Validation of Deep Convolutional Generative Adversarial Networks for High Energy Physics Calorimeter Simulations

In particle physics the simulation of particle transport through detectors requires an enormous amount of computational resources, utilizing more than 50% of the resources of the CERN Worldwide Large Hadron Collider Grid. This challenge has motivated the investigation of different, faster approaches...

Descripción completa

Detalles Bibliográficos
Autores principales: Rehm, Florian, Vallecorsa, Sofia, Borras, Kerstin, Krücker, Dirk
Lenguaje:eng
Publicado: 2021
Materias:
Acceso en línea:http://cds.cern.ch/record/2759294
_version_ 1780970218158292992
author Rehm, Florian
Vallecorsa, Sofia
Borras, Kerstin
Krücker, Dirk
author_facet Rehm, Florian
Vallecorsa, Sofia
Borras, Kerstin
Krücker, Dirk
author_sort Rehm, Florian
collection CERN
description In particle physics the simulation of particle transport through detectors requires an enormous amount of computational resources, utilizing more than 50% of the resources of the CERN Worldwide Large Hadron Collider Grid. This challenge has motivated the investigation of different, faster approaches for replacing the standard Monte Carlo simulations. Deep Learning Generative Adversarial Networks are among the most promising alternatives. Previous studies showed that they achieve the necessary level of accuracy while decreasing the simulation time by orders of magnitudes. In this paper we present a newly developed neural network architecture which reproduces a three-dimensional problem employing 2D convolutional layers and we compare its performance with an earlier architecture consisting of 3D convolutional layers. The performance evaluation relies on direct comparison to Monte Carlo simulations, in terms of different physics quantities usually employed to quantify the detector response. We prove that our new neural network architecture reaches a higher level of accuracy with respect to the 3D convolutional GAN while reducing the necessary computational resources. Calorimeters are among the most expensive detectors in terms of simulation time. Therefore we focus our study on an electromagnetic calorimeter prototype with a regular highly granular geometry, as an example of future calorimeters.
id cern-2759294
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2021
record_format invenio
spelling cern-27592942023-01-31T03:57:44Zhttp://cds.cern.ch/record/2759294engRehm, FlorianVallecorsa, SofiaBorras, KerstinKrücker, DirkValidation of Deep Convolutional Generative Adversarial Networks for High Energy Physics Calorimeter Simulationshep-exParticle Physics - ExperimentIn particle physics the simulation of particle transport through detectors requires an enormous amount of computational resources, utilizing more than 50% of the resources of the CERN Worldwide Large Hadron Collider Grid. This challenge has motivated the investigation of different, faster approaches for replacing the standard Monte Carlo simulations. Deep Learning Generative Adversarial Networks are among the most promising alternatives. Previous studies showed that they achieve the necessary level of accuracy while decreasing the simulation time by orders of magnitudes. In this paper we present a newly developed neural network architecture which reproduces a three-dimensional problem employing 2D convolutional layers and we compare its performance with an earlier architecture consisting of 3D convolutional layers. The performance evaluation relies on direct comparison to Monte Carlo simulations, in terms of different physics quantities usually employed to quantify the detector response. We prove that our new neural network architecture reaches a higher level of accuracy with respect to the 3D convolutional GAN while reducing the necessary computational resources. Calorimeters are among the most expensive detectors in terms of simulation time. Therefore we focus our study on an electromagnetic calorimeter prototype with a regular highly granular geometry, as an example of future calorimeters.arXiv:2103.13698oai:cds.cern.ch:27592942021-03-25
spellingShingle hep-ex
Particle Physics - Experiment
Rehm, Florian
Vallecorsa, Sofia
Borras, Kerstin
Krücker, Dirk
Validation of Deep Convolutional Generative Adversarial Networks for High Energy Physics Calorimeter Simulations
title Validation of Deep Convolutional Generative Adversarial Networks for High Energy Physics Calorimeter Simulations
title_full Validation of Deep Convolutional Generative Adversarial Networks for High Energy Physics Calorimeter Simulations
title_fullStr Validation of Deep Convolutional Generative Adversarial Networks for High Energy Physics Calorimeter Simulations
title_full_unstemmed Validation of Deep Convolutional Generative Adversarial Networks for High Energy Physics Calorimeter Simulations
title_short Validation of Deep Convolutional Generative Adversarial Networks for High Energy Physics Calorimeter Simulations
title_sort validation of deep convolutional generative adversarial networks for high energy physics calorimeter simulations
topic hep-ex
Particle Physics - Experiment
url http://cds.cern.ch/record/2759294
work_keys_str_mv AT rehmflorian validationofdeepconvolutionalgenerativeadversarialnetworksforhighenergyphysicscalorimetersimulations
AT vallecorsasofia validationofdeepconvolutionalgenerativeadversarialnetworksforhighenergyphysicscalorimetersimulations
AT borraskerstin validationofdeepconvolutionalgenerativeadversarialnetworksforhighenergyphysicscalorimetersimulations
AT kruckerdirk validationofdeepconvolutionalgenerativeadversarialnetworksforhighenergyphysicscalorimetersimulations