Göttingen 2025 – scientific programme

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

T: Fachverband Teilchenphysik

T 75: Neutrino Astronomy IV

T 75.5: Talk

Thursday, April 3, 2025, 17:15–17:30, VG 1.105

Advanced Northern Tracks Selection using a Graph Convolutional Neural Network for the IceCube Neutrino Observatory: Event classification — •Shuyang Deng, Philipp Behrens, Jakob Böttcher, Lasse Düse, Philipp Fürst, Leon Hamacher, Michael Handt, Lars Marten, Philipp Soldin, and Christopher Wiebusch for the IceCube collaboration — III. Physikalisches Institut B RWTH Aachen University, Aachen, Germany

The IceCube Neutrino Observatory is a large neutrino detector built deep in the Antarctic ice at the South Pole. The Advanced Northern Track Selection (ANTS) framework uses a graph convolutional neural network to select neutrino-induced muon tracks. These events have different topologies and signatures within the detector, such as through-going, starting, skimming tracks as well as remaining background from cosmic-ray induced muons and cascades e.g. related to electron neutrinos. In this work, we perform the classification of these event topologies using the ANTS framework, providing an event-wise probability for each topology. This classification enables dedicated handling of these topologies in further analyses.

Keywords: IceCube; Neutrino Astronomy; Deep Learning; Advanced Northern Track Selection; Event Selection