Göttingen 2025 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

T: Fachverband Teilchenphysik

T 83: Methods in Particle Physics IV (Lepton Reconstruction)

T 83.10: Talk

Thursday, April 3, 2025, 18:30–18:45, VG 4.101

Estimation of Non-Prompt Lepton Backgrounds with Classical and Machine Learning Techniques — Korn Steffen, Quadt Arnulf, and •Schiel Nico — II. Physikalisches Institut, Georg- August-Universität Göttingen, Friedrich-Hund-Platz 1, 37077 Göttingen

Non-prompt leptons are a significant background in many particle physics analyses, for example tt and HWW^* analyses. These processes depend on the modelling of parton showers and are therefore challenging to predict theoretically. Consequently, data-driven approaches are utilised to model backgrounds arising from non-prompt leptons. Often, classical methods such as the fake-factor method are used. However, machine learning based methods such as normalising flows also show promising results for modelling non-prompt leptons. In this talk, both approaches are compared with respect to their performance.

Keywords: Fake lepton estimation; Machine learning techniques; Normalising Flows