Studies of Gamma Ray Shower Reconstruction Using Deep Learning

Abstract:
'The ALTO project aims to build a particle detector array for very high energy gamma ray observations optimized for soft spectrum sources. The accurate reconstruction of gamma ray events, in particular their energies, using a surface array is an especially challenging problem at the low energies ALTO aims to optimize for. In this contribution, we leverage Convolutional Neural Networks (CNNs) to improve reconstruction performance at lower energies ( smaller 1 TeV ) as compared to the SEMLA analysis procedure, which is a more traditional method using mainly manually derived features.rnWe present performance figures using different network architectures and training settings, both in terms of accuracy and training time, as well as the impact of various data augmentation techniques.'

Authors: Tomas Bylund
Co-Authors: Gašper Kukec Mezek | Mohanraj Senniappan | Yvonne Becherini | Michael Punch
Indico-ID: 1204
Proceeding URL: https://pos.sissa.it/395/758

Discussion timeslot (ZOOM-Meeting): 13. July 2021 - 12:00
ZOOM-Meeting URL: https://desy.zoom.us/j/98542982538
ZOOM-Meeting ID: 98542982538
ZOOM-Meeting Passcode: ICRC2021
Corresponding Session: https://icrc2021-venue.desy.de/channel/52-Analysis-Methods-Catalogues-Community-Tools-Machine-Learning-GAD-GAI/64
Live-Stream URL: https://icrc2021-venue.desy.de/livestream/Discussion-04/5