Würzburg 2018 – wissenschaftliches Programm

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

T: Fachverband Teilchenphysik

T 87: Datenanalyse

T 87.8: Vortrag

Donnerstag, 22. März 2018, 18:15–18:30, Z6 - SR 2.005

Studies for Top Quark Reconstruction with Deep Learning — •Tim Kallage, Johannes Erdmann, Olaf Nackenhorst, and Kevin Kröninger — TU Dortmund, Experimentelle Physik IV

Deep learning techniques are attracting attention in recent years and show potential in high energy physics applications. In analyses of tt processes, a reconstruction of the association of measured jets to partons in the decay topology is often useful. A deep neural network approach for this goal is presented in this talk for semileptonic tt decays. The algorithm is trained and tested on pp collisions at √s = 13 TeV using a simplified simulation of the ATLAS detector. The performance is studied and compared with a commonly used kinematic likelihood fit (KLFitter).