Dortmund 2021 – scientific programme

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

T: Fachverband Teilchenphysik

T 52: Top quark decay and top properties II

T 52.6: Talk

Wednesday, March 17, 2021, 17:15–17:30, Tb

Neural network optimisation in the search for flavour-changing neutral currents in processes with a top quark and a photon at the ATLAS experiment — Tomas Dado, Johannes Erdmann, Benedikt Gocke, •Florian Mausolf, and Olaf Nackenhorst — Experimentelle Physik IV, TU Dortmund, Germany

A search for flavour-changing neutral currents (FCNC) in processes with a top quark and a photon is presented. In the Standard Model (SM) of particle physics these processes are highly suppressed, so that an observation would be a direct hint to physics beyond the SM. Proton-proton collision data taken with the ATLAS detector at a centre-of-mass energy of √s = 13 TeV are used. The signal region targets two possible signal modes, the FCNC production of a single top quark and the decay of a top quark via an FCNC process.

A neural network is used to maximise the separation between signal events and the SM background. Two different network architectures are tested: A binary classifier that discriminates between signal and background, and a multiclass network that evaluates if events originate from the production mode, the decay mode or a background process. Studies for the optimisation of the neural networks as well as the resulting expected upper limits on the signal strength are presented.