Isolating Unisolated Upsilons with Anomaly Detection in CMS Open Data
Physical Review Letters · 08.07.2025 · 135 (2)
DOI: 10.1103/vvv3-5kkl ↗Kurz gefasst
Fachartikel in der Zeitschrift Physical Review Letters, online erschienen am 08.07.2025. Verfasst von Rikab Gambhir und 3 weiteren Autor:innen. Beteiligte Einrichtung: Massachusetts Institute of Technology, Center for Theoretical Physics -- a Leinweber Institute, Cambridge, Massachusetts 02139, USA. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science Foundation, U.S. Department of Energy, High Energy Physics.
Abstract (Originalsprache)
We present the first study of anti-isolated Upsilon decays to two muons ( ϒ → μ + μ − ) in proton-proton collisions at the Large Hadron Collider. Using a machine learning (ML)-based anomaly detection strategy, we “rediscover” the ϒ in 13 TeV CMS Open Data from 2016, despite overwhelming anti-isolated backgrounds. We elevate the signal significance to 6.4 σ using these methods, starting from 1.6 σ using the dimuon mass spectrum alone. Moreover, we demonstrate improved sensitivity from using an ML-based estimate of the multifeature likelihood compared to traditional “cut-and-count” methods. This is the first ever detection of anti-isolated Upsilons, which can be useful in the study of heavy-flavor fragmentation in quantum chromodynamics. Our Letter demonstrates that it is possible and practical to find real signals in experimental collider data using ML-based anomaly detection, and we distill a readily accessible benchmark dataset from the CMS Open Data to facilitate future anomaly detection developments.
Autor:innen und Institutionen
- Massachusetts Institute of Technology, Center for Theoretical Physics -- a Leinweber Institute, Cambridge, Massachusetts 02139, USA.
- The NSF AI Institute for Artificial Intelligence and Fundamental Interactions, USA.
- University of California, Department of Physics, Berkeley, California 94720, USA.
- Lawrence Berkeley National Laboratory, Physics Division, Berkeley, California 94720, USA.
- University of California, Berkeley Institute for Data Science, Berkeley, California 94720, USA.
- Stanford University, Department of Particle Physics and Astrophysics, Stanford, California 94305, USA.
Förderung
- National Science Foundation
- U.S. Department of Energy
- High Energy Physics
- Simons Foundation
- Kavli Institute for Theoretical Physics, University of California, Santa Barbara
- Office of Science
- John Templeton Foundation
- National Energy Research Scientific Computing Center
Zitieren
APA
Gambhir, R., Mastandrea, R., Nachman, B., & Thaler, J. (2025). Isolating Unisolated Upsilons with Anomaly Detection in CMS Open Data. Physical Review Letters, 135(2). https://doi.org/10.1103/vvv3-5kkl
Vancouver
Gambhir R, Mastandrea R, Nachman B, Thaler J. Isolating Unisolated Upsilons with Anomaly Detection in CMS Open Data. Physical Review Letters. 2025;135(2). doi:10.1103/vvv3-5kkl
BibTeX
@article{Gambhir2025,
author = {Gambhir, Rikab and Mastandrea, Radha and Nachman, Benjamin and Thaler, Jesse},
title = {Isolating Unisolated Upsilons with Anomaly Detection in CMS Open Data},
journal = {Physical Review Letters},
year = {2025},
volume = {135},
number = {2},
doi = {10.1103/vvv3-5kkl}
}
EUVIMED