Casting a graph net to catch dark showers
Strongly interacting dark sectors predict novel LHC signatures such as semi-visible jets resulting from dark showers that contain both stable and unstable dark mesons. Distinguishing such semi-visible jets from large QCD backgrounds is difficult and constitutes an exciting challenge for jet clas...
Main Author: | |
---|---|
Format: | Article |
Language: | English |
Published: |
SciPost
2021-02-01
|
Series: | SciPost Physics |
Online Access: | https://scipost.org/SciPostPhys.10.2.046 |
id |
doaj-8b8505c862ab49aa982450b5d59e4e1a |
---|---|
record_format |
Article |
spelling |
doaj-8b8505c862ab49aa982450b5d59e4e1a2021-04-19T12:48:03ZengSciPostSciPost Physics2542-46532021-02-0110204610.21468/SciPostPhys.10.2.046Casting a graph net to catch dark showersElias Bernreuther, Thorben Finke, Felix Kahlhoefer, Michael Krämer, Alexander MückStrongly interacting dark sectors predict novel LHC signatures such as semi-visible jets resulting from dark showers that contain both stable and unstable dark mesons. Distinguishing such semi-visible jets from large QCD backgrounds is difficult and constitutes an exciting challenge for jet classification. In this article we explore the potential of supervised deep neural networks to identify semi-visible jets. We show that dynamic graph convolutional neural networks operating on so-called particle clouds outperform convolutional neural networks analysing jet images as well as other neural networks based on Lorentz vectors. We investigate how the performance depends on the properties of the dark shower and discuss training on mixed samples as a strategy to reduce model dependence. By modifying an existing mono-jet analysis we show that LHC sensitivity to dark sectors can be enhanced by more than an order of magnitude by using the dynamic graph network as a dark shower tagger.https://scipost.org/SciPostPhys.10.2.046 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Elias Bernreuther, Thorben Finke, Felix Kahlhoefer, Michael Krämer, Alexander Mück |
spellingShingle |
Elias Bernreuther, Thorben Finke, Felix Kahlhoefer, Michael Krämer, Alexander Mück Casting a graph net to catch dark showers SciPost Physics |
author_facet |
Elias Bernreuther, Thorben Finke, Felix Kahlhoefer, Michael Krämer, Alexander Mück |
author_sort |
Elias Bernreuther, Thorben Finke, Felix Kahlhoefer, Michael Krämer, Alexander Mück |
title |
Casting a graph net to catch dark showers |
title_short |
Casting a graph net to catch dark showers |
title_full |
Casting a graph net to catch dark showers |
title_fullStr |
Casting a graph net to catch dark showers |
title_full_unstemmed |
Casting a graph net to catch dark showers |
title_sort |
casting a graph net to catch dark showers |
publisher |
SciPost |
series |
SciPost Physics |
issn |
2542-4653 |
publishDate |
2021-02-01 |
description |
Strongly interacting dark sectors predict novel LHC signatures such as
semi-visible jets resulting from dark showers that contain both stable and
unstable dark mesons. Distinguishing such semi-visible jets from large QCD
backgrounds is difficult and constitutes an exciting challenge for jet
classification. In this article we explore the potential of supervised deep
neural networks to identify semi-visible jets. We show that dynamic graph
convolutional neural networks operating on so-called particle clouds outperform
convolutional neural networks analysing jet images as well as other neural
networks based on Lorentz vectors. We investigate how the performance depends
on the properties of the dark shower and discuss training on mixed samples as a
strategy to reduce model dependence. By modifying an existing mono-jet analysis
we show that LHC sensitivity to dark sectors can be enhanced by more than an
order of magnitude by using the dynamic graph network as a dark shower tagger. |
url |
https://scipost.org/SciPostPhys.10.2.046 |
work_keys_str_mv |
AT eliasbernreutherthorbenfinkefelixkahlhoefermichaelkrameralexandermuck castingagraphnettocatchdarkshowers |
_version_ |
1721521185064222720 |