Adversarial Tuning of Perturbative Parameters in Non-Differentiable Physics Simulators

<!--HTML-->In this contribution, we present a method for tuning perturbative parameters in Monte Carlo simulation using a classifier loss in high dimensions. We use an LSTM trained on the radiation pattern inside jets to learn the parameters of the final state shower in the Pythia Monte Carlo...

Descripción completa

Detalles Bibliográficos
Autor principal: Paganini, Michela
Lenguaje:eng
Publicado: 2018
Materias:
Machine Learning
Acceso en línea:http://cds.cern.ch/record/2312417
_version_ 1780957971023396864
author Paganini, Michela
author_facet Paganini, Michela
author_sort Paganini, Michela
collection CERN
description <!--HTML-->In this contribution, we present a method for tuning perturbative parameters in Monte Carlo simulation using a classifier loss in high dimensions. We use an LSTM trained on the radiation pattern inside jets to learn the parameters of the final state shower in the Pythia Monte Carlo generator. This represents a step forward compared to unidimensional distributional template-matching methods.
id cern-2312417
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2018
record_format invenio
spelling cern-23124172022-11-02T22:34:03Zhttp://cds.cern.ch/record/2312417engPaganini, MichelaAdversarial Tuning of Perturbative Parameters in Non-Differentiable Physics Simulators2nd IML Machine Learning WorkshopMachine Learning<!--HTML-->In this contribution, we present a method for tuning perturbative parameters in Monte Carlo simulation using a classifier loss in high dimensions. We use an LSTM trained on the radiation pattern inside jets to learn the parameters of the final state shower in the Pythia Monte Carlo generator. This represents a step forward compared to unidimensional distributional template-matching methods.oai:cds.cern.ch:23124172018
spellingShingle Machine Learning
Paganini, Michela
Adversarial Tuning of Perturbative Parameters in Non-Differentiable Physics Simulators
title Adversarial Tuning of Perturbative Parameters in Non-Differentiable Physics Simulators
title_full Adversarial Tuning of Perturbative Parameters in Non-Differentiable Physics Simulators
title_fullStr Adversarial Tuning of Perturbative Parameters in Non-Differentiable Physics Simulators
title_full_unstemmed Adversarial Tuning of Perturbative Parameters in Non-Differentiable Physics Simulators
title_short Adversarial Tuning of Perturbative Parameters in Non-Differentiable Physics Simulators
title_sort adversarial tuning of perturbative parameters in non-differentiable physics simulators
topic Machine Learning
url http://cds.cern.ch/record/2312417
work_keys_str_mv AT paganinimichela adversarialtuningofperturbativeparametersinnondifferentiablephysicssimulators
AT paganinimichela 2ndimlmachinelearningworkshop

Ejemplares similares