Dortmund 2021 – scientific programme

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

T: Fachverband Teilchenphysik

T 38: Data analysis, information technology II

T 38.5: Talk

Tuesday, March 16, 2021, 17:00–17:15, Tm

CNN classification and regression for ANTARES — •Nicole Geißelbrecht for the ANTARES-KM3NeT-Erlangen collaboration — Friedrich-Alexander-Universität Erlangen-Nürnberg, ECAP

ANTARES is the largest undersea neutrino detector, installed in the Mediterranean Sea, and is primarily sensitive to neutrinos in the TeV-PeV energy range. Data taking with the telescope has been continuous since 2008. One of the central goals of ANTARES, next to searches for neutrino signals from point, transient, and extended sources, is an independent detection and investigation of the diffuse cosmic neutrino flux discovered by IceCube. The suppression of backgrounds, in particular of atmospheric muons, is essential to further increase the sensitivity of the data analysis.

The contribution reports on the design and application of deep Convolutional Neural Networks to ANTARES telescope data. Data preprocessing concepts, image generation, and performance investigations of an event-topology classifier as well as first studies with energy regression will be presented.