Quasi anomalous knowledge: searching for new physics with embedded knowledge
Abstract Discoveries of new phenomena often involve a dedicated search for a hypothetical physics signature. Recently, novel deep learning techniques have emerged for anomaly detection in the absence of a signal prior. However, by ignoring signal priors, the sensitivity of these approaches is signif...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
SpringerOpen
2021-06-01
|
Series: | Journal of High Energy Physics |
Subjects: | |
Online Access: | https://doi.org/10.1007/JHEP06(2021)030 |
id |
doaj-0dd4095870b541f3bababe5e8850f960 |
---|---|
record_format |
Article |
spelling |
doaj-0dd4095870b541f3bababe5e8850f9602021-06-06T11:08:25ZengSpringerOpenJournal of High Energy Physics1029-84792021-06-012021612610.1007/JHEP06(2021)030Quasi anomalous knowledge: searching for new physics with embedded knowledgeSang Eon Park0Dylan Rankin1Silviu-Marian Udrescu2Mikaeel Yunus3Philip Harris4Laboratory for Nuclear Science, Massachusetts Institute of TechnologyLaboratory for Nuclear Science, Massachusetts Institute of TechnologyLaboratory for Nuclear Science, Massachusetts Institute of TechnologyLaboratory for Nuclear Science, Massachusetts Institute of TechnologyLaboratory for Nuclear Science, Massachusetts Institute of TechnologyAbstract Discoveries of new phenomena often involve a dedicated search for a hypothetical physics signature. Recently, novel deep learning techniques have emerged for anomaly detection in the absence of a signal prior. However, by ignoring signal priors, the sensitivity of these approaches is significantly reduced. We present a new strategy dubbed Quasi Anomalous Knowledge (QUAK), whereby we introduce alternative signal priors that capture some of the salient features of new physics signatures, allowing for the recovery of sensitivity even when the alternative signal is incorrect. This approach can be applied to a broad range of physics models and neural network architectures. In this paper, we apply QUAK to anomaly detection of new physics events at the CERN Large Hadron Collider utilizing variational autoencoders with normalizing flow.https://doi.org/10.1007/JHEP06(2021)030Beyond Standard ModelExoticsJet substructureHadron-Hadron scattering (experiments)Jets |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Sang Eon Park Dylan Rankin Silviu-Marian Udrescu Mikaeel Yunus Philip Harris |
spellingShingle |
Sang Eon Park Dylan Rankin Silviu-Marian Udrescu Mikaeel Yunus Philip Harris Quasi anomalous knowledge: searching for new physics with embedded knowledge Journal of High Energy Physics Beyond Standard Model Exotics Jet substructure Hadron-Hadron scattering (experiments) Jets |
author_facet |
Sang Eon Park Dylan Rankin Silviu-Marian Udrescu Mikaeel Yunus Philip Harris |
author_sort |
Sang Eon Park |
title |
Quasi anomalous knowledge: searching for new physics with embedded knowledge |
title_short |
Quasi anomalous knowledge: searching for new physics with embedded knowledge |
title_full |
Quasi anomalous knowledge: searching for new physics with embedded knowledge |
title_fullStr |
Quasi anomalous knowledge: searching for new physics with embedded knowledge |
title_full_unstemmed |
Quasi anomalous knowledge: searching for new physics with embedded knowledge |
title_sort |
quasi anomalous knowledge: searching for new physics with embedded knowledge |
publisher |
SpringerOpen |
series |
Journal of High Energy Physics |
issn |
1029-8479 |
publishDate |
2021-06-01 |
description |
Abstract Discoveries of new phenomena often involve a dedicated search for a hypothetical physics signature. Recently, novel deep learning techniques have emerged for anomaly detection in the absence of a signal prior. However, by ignoring signal priors, the sensitivity of these approaches is significantly reduced. We present a new strategy dubbed Quasi Anomalous Knowledge (QUAK), whereby we introduce alternative signal priors that capture some of the salient features of new physics signatures, allowing for the recovery of sensitivity even when the alternative signal is incorrect. This approach can be applied to a broad range of physics models and neural network architectures. In this paper, we apply QUAK to anomaly detection of new physics events at the CERN Large Hadron Collider utilizing variational autoencoders with normalizing flow. |
topic |
Beyond Standard Model Exotics Jet substructure Hadron-Hadron scattering (experiments) Jets |
url |
https://doi.org/10.1007/JHEP06(2021)030 |
work_keys_str_mv |
AT sangeonpark quasianomalousknowledgesearchingfornewphysicswithembeddedknowledge AT dylanrankin quasianomalousknowledgesearchingfornewphysicswithembeddedknowledge AT silviumarianudrescu quasianomalousknowledgesearchingfornewphysicswithembeddedknowledge AT mikaeelyunus quasianomalousknowledgesearchingfornewphysicswithembeddedknowledge AT philipharris quasianomalousknowledgesearchingfornewphysicswithembeddedknowledge |
_version_ |
1721394375417659392 |