Cargando…

Fast simulation methods in ATLAS: from classical to generative models

The ATLAS physics program relies on very large samples of \textsc{Geant4} simulated events, which provide a highly detailed and accurate simulation of the ATLAS detector. However, this accuracy comes with a high price in CPU, and the sensitivity of many physics analyses is already limited by the ava...

Descripción completa

Detalles Bibliográficos
Autores principales: Chapman, John Derek, Cranmer, Kyle, Gadatsch, Stefan, Golling, Tobias, Ghosh, Aishik, Gray, Heather, Lari, Tommaso, Pascuzzi, Vincent, Raine, John Andrew, Rousseau, David, Salamani, Dalila, Schaarschmidt, Jana
Lenguaje:eng
Publicado: 2020
Materias:
Acceso en línea:https://dx.doi.org/10.1051/epjconf/202024502035
http://cds.cern.ch/record/2740905
_version_ 1780968369401364480
author Chapman, John Derek
Cranmer, Kyle
Gadatsch, Stefan
Golling, Tobias
Ghosh, Aishik
Gray, Heather
Lari, Tommaso
Pascuzzi, Vincent
Raine, John Andrew
Rousseau, David
Salamani, Dalila
Schaarschmidt, Jana
author_facet Chapman, John Derek
Cranmer, Kyle
Gadatsch, Stefan
Golling, Tobias
Ghosh, Aishik
Gray, Heather
Lari, Tommaso
Pascuzzi, Vincent
Raine, John Andrew
Rousseau, David
Salamani, Dalila
Schaarschmidt, Jana
author_sort Chapman, John Derek
collection CERN
description The ATLAS physics program relies on very large samples of \textsc{Geant4} simulated events, which provide a highly detailed and accurate simulation of the ATLAS detector. However, this accuracy comes with a high price in CPU, and the sensitivity of many physics analyses is already limited by the available Monte Carlo statistics and will be even more so in the future. Therefore, sophisticated fast simulation tools are developed. In Run-3 we aim to replace the calorimeter shower simulation for most samples with a new parametrized description of longitudinal and lateral energy deposits, including machine learning approaches, to achieve a fast and accurate description. Looking further ahead, prototypes are being developed using cutting edge machine learning approaches to learn the appropriate calorimeter response, which are expected to improve modeling of correlations within showers. Two different approaches, using Variational Auto-Encoders (VAEs) or Generative Adversarial Networks (GANs), are trained to model the shower simulation. Additional fast simulation tools will replace the inner detector simulation, as well as digitization and reconstruction algorithms, achieving up to two orders of magnitude improvement in speed. In this talk, we will describe the new tools for fast production of simulated events and an exploratory analysis of the deep learning methods.
id cern-2740905
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2020
record_format invenio
spelling cern-27409052021-03-08T20:51:51Zdoi:10.1051/epjconf/202024502035http://cds.cern.ch/record/2740905engChapman, John DerekCranmer, KyleGadatsch, StefanGolling, TobiasGhosh, AishikGray, HeatherLari, TommasoPascuzzi, VincentRaine, John AndrewRousseau, DavidSalamani, DalilaSchaarschmidt, JanaFast simulation methods in ATLAS: from classical to generative modelsParticle Physics - ExperimentThe ATLAS physics program relies on very large samples of \textsc{Geant4} simulated events, which provide a highly detailed and accurate simulation of the ATLAS detector. However, this accuracy comes with a high price in CPU, and the sensitivity of many physics analyses is already limited by the available Monte Carlo statistics and will be even more so in the future. Therefore, sophisticated fast simulation tools are developed. In Run-3 we aim to replace the calorimeter shower simulation for most samples with a new parametrized description of longitudinal and lateral energy deposits, including machine learning approaches, to achieve a fast and accurate description. Looking further ahead, prototypes are being developed using cutting edge machine learning approaches to learn the appropriate calorimeter response, which are expected to improve modeling of correlations within showers. Two different approaches, using Variational Auto-Encoders (VAEs) or Generative Adversarial Networks (GANs), are trained to model the shower simulation. Additional fast simulation tools will replace the inner detector simulation, as well as digitization and reconstruction algorithms, achieving up to two orders of magnitude improvement in speed. In this talk, we will describe the new tools for fast production of simulated events and an exploratory analysis of the deep learning methods.ATL-SOFT-PROC-2020-035oai:cds.cern.ch:27409052020-10-08
spellingShingle Particle Physics - Experiment
Chapman, John Derek
Cranmer, Kyle
Gadatsch, Stefan
Golling, Tobias
Ghosh, Aishik
Gray, Heather
Lari, Tommaso
Pascuzzi, Vincent
Raine, John Andrew
Rousseau, David
Salamani, Dalila
Schaarschmidt, Jana
Fast simulation methods in ATLAS: from classical to generative models
title Fast simulation methods in ATLAS: from classical to generative models
title_full Fast simulation methods in ATLAS: from classical to generative models
title_fullStr Fast simulation methods in ATLAS: from classical to generative models
title_full_unstemmed Fast simulation methods in ATLAS: from classical to generative models
title_short Fast simulation methods in ATLAS: from classical to generative models
title_sort fast simulation methods in atlas: from classical to generative models
topic Particle Physics - Experiment
url https://dx.doi.org/10.1051/epjconf/202024502035
http://cds.cern.ch/record/2740905
work_keys_str_mv AT chapmanjohnderek fastsimulationmethodsinatlasfromclassicaltogenerativemodels
AT cranmerkyle fastsimulationmethodsinatlasfromclassicaltogenerativemodels
AT gadatschstefan fastsimulationmethodsinatlasfromclassicaltogenerativemodels
AT gollingtobias fastsimulationmethodsinatlasfromclassicaltogenerativemodels
AT ghoshaishik fastsimulationmethodsinatlasfromclassicaltogenerativemodels
AT grayheather fastsimulationmethodsinatlasfromclassicaltogenerativemodels
AT laritommaso fastsimulationmethodsinatlasfromclassicaltogenerativemodels
AT pascuzzivincent fastsimulationmethodsinatlasfromclassicaltogenerativemodels
AT rainejohnandrew fastsimulationmethodsinatlasfromclassicaltogenerativemodels
AT rousseaudavid fastsimulationmethodsinatlasfromclassicaltogenerativemodels
AT salamanidalila fastsimulationmethodsinatlasfromclassicaltogenerativemodels
AT schaarschmidtjana fastsimulationmethodsinatlasfromclassicaltogenerativemodels