Karlsruhe 2024 – scientific programme

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

T: Fachverband Teilchenphysik

T 43: Data, AI, Computing 3 (pointclouds & graphs)

T 43.5: Talk

Tuesday, March 5, 2024, 17:00–17:15, Geb. 30.33: MTI

Search for fractionally charged particles with Graph Neural Networks — •Alexander Sandrock and Timo Stürwald — Bergische Universität Wuppertal, Wuppertal, Deutschland

Fractionally charged particles are hypothetical particles with a charge smaller than the electron charge. These particles are predicted in various theories Beyond the Standard Model of particle physics, for instance in versions of supersymmetry. The IceCube neutrino observatory as a very large volume detector shielded by a kilometer-thick ice shield is ideally suited to search for signatures of rare particles such as fractionally charged particles.

Graph neural networks have been successfully applied in the last few years to the classification and reconstruction of events in the IceCube detector. This presentation discusses the application of graph neural networks to the discrimination between simulated fractionally charged particle events and standard model simulations.

Keywords: IceCube; fractional charges; graph neural networks