Energy Reconstruction of Non-fiducial Electron-Positron Events in the DAMPE Experiment Using Convolutional Neural Networks

Abstract

The Dark Matter Particle Explorer (DAMPE) is a space-based Cosmic-Ray (CR) observatory with the aim, among others, to study Cosmic-Ray Electrons (CREs) up to 10 TeV. Due to the low CRE rate at multi-TeV energies, we aim to increasing the acceptance by selecting events outside the fiducial volume. The complex topology of non-fiducial events requires the development of a novel energy reconstruction method. We propose the usage of Convolutional Neural Networks for a regression task to recover an accurate estimation of the initial energy.