Erlangen 2026 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

T: Fachverband Teilchenphysik

T 28: Data, AI, Computing, Electronics III

T 28.6: Vortrag

Dienstag, 17. März 2026, 17:30–17:45, KH 00.024

Machine-Learning Based Reconstruction of Muon Detector Showers in CMS — •Ayse Asu Guvenli, Karim El Morabit, Gregor Kasieczka, and Mascha Hackmann — University of Hamburg, Hamburg, Germany

Long-lived particles (LLPs) appear in many extensions of the Standard Model and can travel measurable distances before decaying. When such decays occur in or near the CMS muon system, they can produce atypical showers of hits rather than standard muon signatures. These muon detector showers (MDS) offer a promising handle for LLP searches but remain difficult to reconstruct with algorithms optimized for muons.

This work explores machine-learning based reconstruction of shower-like hit patterns in the CMS Cathode Strip Chambers, aiming to improve the identification and characterization of MDS from LLP decays. We present ongoing developments toward low-level hit reconstruction with ML methods and discuss their potential to enhance the sensitivity of Run-3 LLP searches.

Keywords: Long-lived particles; Muon detector showers; Machine learning; Hit reconstruction; CMS detector