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

T: Fachverband Teilchenphysik

T 59: Neutrino Astronomy III

T 59.8: Vortrag

Mittwoch, 18. März 2026, 18:00–18:15, KS H C

Neural Network Tools for the IceCube-Gen2 Optical Array — •Francisco Javier Vara Carbonell and Alexander Kappes for the IceCube collaboration — Universität Münster, Institut für Kernphysik

IceCube-Gen2 is a planned extension of the current IceCube neutrino observatory that will increase the in-ice instrumented volume by roughly a factor of eight and introduce 9600 new optical sensors with multiple photomultiplier tubes (PMTs) designed to provide nearly full solid-angle coverage. In this new configuration, neural networks are particularly strong candidates to replace or complement traditional algorithms in key tasks such as event reconstruction, topology classification, noise cleaning, and simulation, as they naturally exploit the additional information provided by the new sensors and can process large data volumes efficiently, especially when run on graphics processing units (GPUs). This talk will provide an overview of these methods, summarize their current performance in the IceCube-Gen2 context, and discuss their possible physics implications.

Keywords: IceCube-Gen2; Neural Networks; Simulation; Reconstruction