# Developing Neural Networks to Tag Emerging Jets in the ATLAS Experiment

- Institution: CERN
- Department: Summer Student Lecture Programme General
- Speaker: Sophie Gélinas
- Year: 2025
- Language: English
- Source: [CERN](https://videos.cern.ch/record/3023154)
- Licence: [CERN](http://copyright.web.cern.ch)
- Who can watch: Teens profiles and older (sources give no ratings; placed by Invisible Network from the source's own words)

Machine learning algorithms are being developed to search for dark matter in the ATLAS experiment. One class of dark matter models under consideration involves a dark sector containing dark quarks that shower and hadronize into dark mesons. These mesons may travel significant distances before decaying into Standard Model particles, producing topologically distinct emerging jets that can be identified by neural networks. This talk will present improvements to an existing supervised jet tagger, as well as the development of a new unsupervised algorithm aimed at enabling model-independent searches.

Sophie is an undergraduate physics student at the University of British Columbia in Vancouver, Canada. This summer, she is working with the ATLAS group at Simon Fraser University..

0:00:00 Slide 1

0:01:50 Slide 2

0:02:59 Slide 3

0:07:31 Slide 4

0:08:39 Slide 5

0:10:08 Slide 6

0:11:41 Slide 7

## Lectures

1. [Developing Neural Networks to Tag Emerging Jets in the ATLAS Experiment](https://invisiblenetwork.ai/learn/cern/3023154/308e6baab6a1), 13 min (CERN's player)