Cargando…

Machine Learning with ROOT/TMVA

ROOT provides, through TMVA, machine learning tools for data analysis at HEP experiments and beyond. We present recently included features in TMVA and the strategy for future developments in the diversified machine learning landscape. Focus is put on fast machine learning inference, which enables an...

Descripción completa

Detalles Bibliográficos
Autores principales: Albertsson, Kim, An, Sitong, Gleyzer, Sergei, Moneta, Lorenzo, Niermann, Joana, Wunsch, Stefan, Zampieri, Luca, Mesa, Omar Andres Zapata
Lenguaje:eng
Publicado: 2020
Materias:
Acceso en línea:https://dx.doi.org/10.1051/epjconf/202024506019
http://cds.cern.ch/record/2752303
_version_ 1780969334683729920
author Albertsson, Kim
An, Sitong
Gleyzer, Sergei
Moneta, Lorenzo
Niermann, Joana
Wunsch, Stefan
Zampieri, Luca
Mesa, Omar Andres Zapata
author_facet Albertsson, Kim
An, Sitong
Gleyzer, Sergei
Moneta, Lorenzo
Niermann, Joana
Wunsch, Stefan
Zampieri, Luca
Mesa, Omar Andres Zapata
author_sort Albertsson, Kim
collection CERN
description ROOT provides, through TMVA, machine learning tools for data analysis at HEP experiments and beyond. We present recently included features in TMVA and the strategy for future developments in the diversified machine learning landscape. Focus is put on fast machine learning inference, which enables analysts to deploy their machine learning models rapidly on large scale datasets. The new developments are paired with newly designed C++ and Python interfaces supporting modern C++ paradigms and full interoperability in the Python ecosystem. We present as well a new deep learning implementation for convolutional neural network using the cuDNN library for GPU. We show benchmarking results in term of training time and inference time, when comparing with other machine learning libraries such as Keras/Tensorflow.
id oai-inspirehep.net-1832217
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2020
record_format invenio
spelling oai-inspirehep.net-18322172021-02-18T19:23:11Zdoi:10.1051/epjconf/202024506019http://cds.cern.ch/record/2752303engAlbertsson, KimAn, SitongGleyzer, SergeiMoneta, LorenzoNiermann, JoanaWunsch, StefanZampieri, LucaMesa, Omar Andres ZapataMachine Learning with ROOT/TMVAComputing and ComputersROOT provides, through TMVA, machine learning tools for data analysis at HEP experiments and beyond. We present recently included features in TMVA and the strategy for future developments in the diversified machine learning landscape. Focus is put on fast machine learning inference, which enables analysts to deploy their machine learning models rapidly on large scale datasets. The new developments are paired with newly designed C++ and Python interfaces supporting modern C++ paradigms and full interoperability in the Python ecosystem. We present as well a new deep learning implementation for convolutional neural network using the cuDNN library for GPU. We show benchmarking results in term of training time and inference time, when comparing with other machine learning libraries such as Keras/Tensorflow.oai:inspirehep.net:18322172020
spellingShingle Computing and Computers
Albertsson, Kim
An, Sitong
Gleyzer, Sergei
Moneta, Lorenzo
Niermann, Joana
Wunsch, Stefan
Zampieri, Luca
Mesa, Omar Andres Zapata
Machine Learning with ROOT/TMVA
title Machine Learning with ROOT/TMVA
title_full Machine Learning with ROOT/TMVA
title_fullStr Machine Learning with ROOT/TMVA
title_full_unstemmed Machine Learning with ROOT/TMVA
title_short Machine Learning with ROOT/TMVA
title_sort machine learning with root/tmva
topic Computing and Computers
url https://dx.doi.org/10.1051/epjconf/202024506019
http://cds.cern.ch/record/2752303
work_keys_str_mv AT albertssonkim machinelearningwithroottmva
AT ansitong machinelearningwithroottmva
AT gleyzersergei machinelearningwithroottmva
AT monetalorenzo machinelearningwithroottmva
AT niermannjoana machinelearningwithroottmva
AT wunschstefan machinelearningwithroottmva
AT zampieriluca machinelearningwithroottmva
AT mesaomarandreszapata machinelearningwithroottmva