Identifying the relevant dependencies of the neural network response on characteristics of the input space

Mostrar otras versiones (2)

<!--HTML-->The use of neural networks in physics analyses poses new challenges for the estimation of systematic uncertainties. Since the key to a proper estimation of uncertainties is the precise understanding of the algorithm, novel methods for the detailed study of the trained neural network...

Descripción completa

Detalles Bibliográficos
Autor principal: Wunsch, Stefan
Lenguaje:eng
Publicado: 2018
Materias:
Machine Learning
Acceso en línea:http://cds.cern.ch/record/2312450
_version_ 1780957973567242240
author Wunsch, Stefan
author_facet Wunsch, Stefan
author_sort Wunsch, Stefan
collection CERN
description <!--HTML-->The use of neural networks in physics analyses poses new challenges for the estimation of systematic uncertainties. Since the key to a proper estimation of uncertainties is the precise understanding of the algorithm, novel methods for the detailed study of the trained neural network are valuable. This talk presents an approach to identify those characteristics of the neural network inputs that are most relevant for the response and therefore provides essential information to determine the systematic uncertainties.
id cern-2312450
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2018
record_format invenio
spelling cern-23124502022-11-02T22:34:03Zhttp://cds.cern.ch/record/2312450engWunsch, StefanIdentifying the relevant dependencies of the neural network response on characteristics of the input space2nd IML Machine Learning WorkshopMachine Learning<!--HTML-->The use of neural networks in physics analyses poses new challenges for the estimation of systematic uncertainties. Since the key to a proper estimation of uncertainties is the precise understanding of the algorithm, novel methods for the detailed study of the trained neural network are valuable. This talk presents an approach to identify those characteristics of the neural network inputs that are most relevant for the response and therefore provides essential information to determine the systematic uncertainties.oai:cds.cern.ch:23124502018
spellingShingle Machine Learning
Wunsch, Stefan
Identifying the relevant dependencies of the neural network response on characteristics of the input space
title Identifying the relevant dependencies of the neural network response on characteristics of the input space
title_full Identifying the relevant dependencies of the neural network response on characteristics of the input space
title_fullStr Identifying the relevant dependencies of the neural network response on characteristics of the input space
title_full_unstemmed Identifying the relevant dependencies of the neural network response on characteristics of the input space
title_short Identifying the relevant dependencies of the neural network response on characteristics of the input space
title_sort identifying the relevant dependencies of the neural network response on characteristics of the input space
topic Machine Learning
url http://cds.cern.ch/record/2312450
work_keys_str_mv AT wunschstefan identifyingtherelevantdependenciesoftheneuralnetworkresponseoncharacteristicsoftheinputspace
AT wunschstefan 2ndimlmachinelearningworkshop

Ejemplares similares