Erlangen 2026 – wissenschaftliches Programm

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

T: Fachverband Teilchenphysik

T 46: Top Physics II

T 46.5: Vortrag

Mittwoch, 18. März 2026, 17:15–17:30, KH 00.011

Using Neural Networks to Identify Pure Signal Regions for tqγ Production at the ATLAS Experiment — •Marina Andreß¹, Andrea Knue², and Lucas Cremer² — ¹Bergische Universität Wuppertal — ²TU Dortmund

Following the observation of single-top production in association with a photon at the ATLAS experiment, a differential cross-section measurement is performed. An event classification strategy for tqγ events is developed, with the goal of defining a pure signal region suitable for a subsequent unfolding measurement. Such a region is essential to enable differential studies of this rare process, which directly probe the electroweak coupling of the top quark. Therefore, two machine learning approaches are investigated: a deep neural network with a conventional feed-forward structure and a graph neural network that incorporates event topology. The results are estimated using the full ATLAS Run-2 dataset, corresponding to an integrated luminosity of 140 fb⁻¹.

Keywords: single top; Neural Network; Run 2