Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
T: Fachverband Teilchenphysik
T 8: Top Physics I (tt+X)
T 8.6: Vortrag
Montag, 31. März 2025, 18:00–18:15, VG 1.103
Development of General Purpose tt + X and t + X Classifiers with ATLAS — •Clinton Gonsalves, Steffen Korn, and Arnulf Quadt — I Physikalisches Institut, Georg-August-Universität Goettingen, Friedrich-Hund-Platz 1, 37077 Goettingen
The study of tt + X and t + X processes is important for understanding the Standard Model (SM) and exploring potential new physics beyond it. These processes involve the production of a top quark pair ( tt ) or a single top quark ( t ) in association with an additional boson, such as a W , Z , Higgs boson ( H ), or a photon ( γ ). Accurately identifying these events is crucial for measuring properties such as the top Yukawa coupling in tt + H production and for studying rare electroweak processes such as associated t−Z production. However, the rarity of these events and their similarity to background processes, such as tt + jets , make their classification challenging.
This talk presents the development of a neural network-based classifier for tt + X and t + X events. The classifier is trained on kinematic variables, event-level observables such as transverse momentum pTMiss and missing transverse energy ETMiss , and object counts, including jets, leptons, and b -tagged jets. The neural network architecture is optimised to effectively distinguish signal events from background events by learning correlations in the input features.
Keywords: Top Quark; Experimental Particle Physics; Standard Model Particle Physics; Neural Networks In Particle Physics; ATLAS Detector