Speakers
Description
The KM3NeT next generation deep-sea neutrino telescopes are currently under construction in the Mediterranean Sea. Two water-Cherenkov neutrino detectors, ARCA and ORCA, are located in two different sites, south-est of Portopalo di Capopassero (Italy) and close to Toulon (France), respectively. The KM3NeT/ARCA telescope, a cubic kilometer volume detector, is optimised for the detection of high-energy astrophysical neutrinos in the TeV-PeV range. Once completed, the detector will consist of 230 Detection Units, each housing 18 Optical Modules. In order to search for neutrino signals, a high background rejection power is needed and deep learning techniques provide promising methods for achieving this result. The flexibility of the so-called Graph Neural Networks (GNNs) suits perfectly the topology of a complex detector such as KM3NeT.
This contribution will be focused on two interesting applications of GNNs: discrimination of signal events from the background, mainly composed of atmospheric induced events, and energy and direction event reconstruction.
| Poster prize | Yes |
|---|---|
| Collaboration (if any) | KM3NET |
