Speaker
Hugo Boutin
Description
The Dark Matter Particle Explorer experiment allows for γ-ray detection up to TeV energies, with an unprecedented energy resolution of about 1%, which makes it a unique instrument for
γ-ray physics at these energies. A deep-learning tool for track reconstruction has already been developed for electrons and ions. We used this tool on γ-ray samples to assess its efficiency on trajectory reconstruction up to 10 TeV. Preliminary results using a deep-learning model trained on electron samples and applied to γ-ray samples already show very promising results and bring the prospect of this new tool in high energy γ-ray study. The efficiency of the γ selection by the deep-learning method is compared to the classical Kalman-filter-based techniques.
Primary authors
Hugo Boutin
Andrii Tykhonov
(University of Geneva)
Prof.
Xin Wu
(University of Geneva)
Jennifer Frieden
Dr
Chiara Perrina
(EPFL)
