Die DPG-Frühjahrstagung in Bonn musste abgesagt werden! Lesen Sie mehr ...

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

T: Fachverband Teilchenphysik

T 27: Higgs: Decay into fermions I

T 27.4: Vortrag

Dienstag, 31. März 2020, 17:45–18:00, H-HS X

Mass reconstruction with neural networks for a lepton-flavour violating Higgs boson with the ATLAS experiment at √s=13 TeV — •Emanuel Dorbath, Valerie Lang, Katharina Schleicher, and Markus Schumacher — Albert-Ludwigs-Universität Freiburg

The discovery of the Higgs boson allows to search for lepton-flavour violating (LFV) processes in the Higgs-boson sector. Many extensions of the standard model predict this violation, for instance supersymmetric extensions. The general existence of LFV processes in nature has been demonstrated with the observation of neutrino oscillations.

At the ATLAS experiment, interesting LFV Higgs decays are H→ eτ and H→µτ. Leptonic tau-lepton decays are considered yielding the final state eµ2ν. Both neutrinos leave the ATLAS detector without detection, making the reconstruction of the Higgs-boson mass challenging. Non-detected particles broaden the mass resolution and thus complicate the separation of the LFV signal from standard model background processes. Improving the resolution of the mass reconstruction will therefore increase the sensitivity to small branching ratios for LFV decays.

A deep neuronal network is trained in order to reconstruct the mass of a spin 0 boson decaying to τ e→ eµ2ν. The talk describes the optimization of the network architecture and training process. The results in terms of bias and resolution will be compared to standard methods for mass reconstruction.