Semi-visible jets, energy-based models, and self-supervision
We present DarkCLR, a novel framework for detecting semi-visible jets at the LHC. DarkCLR uses a self-supervised contrastive-learning approach to create observables that are approximately invariant under relevant transformations. We use background-enhanced data to create a sensitive representation a...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SciPost
2025-02-01
|
| Series: | SciPost Physics |
| Online Access: | https://scipost.org/SciPostPhys.18.2.042 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832542889633120256 |
|---|---|
| author | Luigi Favaro, Michael Krämer, Tanmoy Modak, Tilman Plehn, Jan Rüschkamp |
| author_facet | Luigi Favaro, Michael Krämer, Tanmoy Modak, Tilman Plehn, Jan Rüschkamp |
| author_sort | Luigi Favaro, Michael Krämer, Tanmoy Modak, Tilman Plehn, Jan Rüschkamp |
| collection | DOAJ |
| description | We present DarkCLR, a novel framework for detecting semi-visible jets at the LHC. DarkCLR uses a self-supervised contrastive-learning approach to create observables that are approximately invariant under relevant transformations. We use background-enhanced data to create a sensitive representation and evaluate the representations using a CLR-inspired anomaly score and a normalized autoencoder as density estimators. Our results show a remarkable sensitivity for a wide range of semi-visible jets and are more robust than a supervised classifier trained on a specific signal. |
| format | Article |
| id | doaj-art-453dc9e32f6b47129c2d7594eacb3f93 |
| institution | Kabale University |
| issn | 2542-4653 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | SciPost |
| record_format | Article |
| series | SciPost Physics |
| spelling | doaj-art-453dc9e32f6b47129c2d7594eacb3f932025-02-03T12:15:07ZengSciPostSciPost Physics2542-46532025-02-0118204210.21468/SciPostPhys.18.2.042Semi-visible jets, energy-based models, and self-supervisionLuigi Favaro, Michael Krämer, Tanmoy Modak, Tilman Plehn, Jan RüschkampWe present DarkCLR, a novel framework for detecting semi-visible jets at the LHC. DarkCLR uses a self-supervised contrastive-learning approach to create observables that are approximately invariant under relevant transformations. We use background-enhanced data to create a sensitive representation and evaluate the representations using a CLR-inspired anomaly score and a normalized autoencoder as density estimators. Our results show a remarkable sensitivity for a wide range of semi-visible jets and are more robust than a supervised classifier trained on a specific signal.https://scipost.org/SciPostPhys.18.2.042 |
| spellingShingle | Luigi Favaro, Michael Krämer, Tanmoy Modak, Tilman Plehn, Jan Rüschkamp Semi-visible jets, energy-based models, and self-supervision SciPost Physics |
| title | Semi-visible jets, energy-based models, and self-supervision |
| title_full | Semi-visible jets, energy-based models, and self-supervision |
| title_fullStr | Semi-visible jets, energy-based models, and self-supervision |
| title_full_unstemmed | Semi-visible jets, energy-based models, and self-supervision |
| title_short | Semi-visible jets, energy-based models, and self-supervision |
| title_sort | semi visible jets energy based models and self supervision |
| url | https://scipost.org/SciPostPhys.18.2.042 |
| work_keys_str_mv | AT luigifavaromichaelkramertanmoymodaktilmanplehnjanruschkamp semivisiblejetsenergybasedmodelsandselfsupervision |