Anomaly Awareness

We present a new algorithm for anomaly detection called Anomaly Awareness. The algorithm learns about normal events while being made aware of the anomalies through a modification of the cost function. We show how this method works in different Particle Physics situations and in standard Computer Vis...

Full description

Bibliographic Details
Main Author: Charanjit K. Khosa, Veronica Sanz
Format: Article
Language:English
Published: SciPost 2023-08-01
Series:SciPost Physics
Online Access:https://scipost.org/SciPostPhys.15.2.053
_version_ 1797751302197870592
author Charanjit K. Khosa, Veronica Sanz
author_facet Charanjit K. Khosa, Veronica Sanz
author_sort Charanjit K. Khosa, Veronica Sanz
collection DOAJ
description We present a new algorithm for anomaly detection called Anomaly Awareness. The algorithm learns about normal events while being made aware of the anomalies through a modification of the cost function. We show how this method works in different Particle Physics situations and in standard Computer Vision tasks. For example, we apply the method to images from a Fat Jet topology generated by Standard Model Top and QCD events, and test it against an array of new physics scenarios, including Higgs production with EFT effects and resonances decaying into two, three or four subjets. We find that the algorithm is effective identifying anomalies not seen before, and becomes robust as we make it aware of a varied-enough set of anomalies.
first_indexed 2024-03-12T16:47:29Z
format Article
id doaj.art-53f92f5bf0d64fbc8caa060266494139
institution Directory Open Access Journal
issn 2542-4653
language English
last_indexed 2024-03-12T16:47:29Z
publishDate 2023-08-01
publisher SciPost
record_format Article
series SciPost Physics
spelling doaj.art-53f92f5bf0d64fbc8caa0602664941392023-08-08T12:01:16ZengSciPostSciPost Physics2542-46532023-08-0115205310.21468/SciPostPhys.15.2.053Anomaly AwarenessCharanjit K. Khosa, Veronica SanzWe present a new algorithm for anomaly detection called Anomaly Awareness. The algorithm learns about normal events while being made aware of the anomalies through a modification of the cost function. We show how this method works in different Particle Physics situations and in standard Computer Vision tasks. For example, we apply the method to images from a Fat Jet topology generated by Standard Model Top and QCD events, and test it against an array of new physics scenarios, including Higgs production with EFT effects and resonances decaying into two, three or four subjets. We find that the algorithm is effective identifying anomalies not seen before, and becomes robust as we make it aware of a varied-enough set of anomalies.https://scipost.org/SciPostPhys.15.2.053
spellingShingle Charanjit K. Khosa, Veronica Sanz
Anomaly Awareness
SciPost Physics
title Anomaly Awareness
title_full Anomaly Awareness
title_fullStr Anomaly Awareness
title_full_unstemmed Anomaly Awareness
title_short Anomaly Awareness
title_sort anomaly awareness
url https://scipost.org/SciPostPhys.15.2.053
work_keys_str_mv AT charanjitkkhosaveronicasanz anomalyawareness