Direct Photon Identification with Artificial Neural Network in the Photon Spectrometer PHOS

Abstract

A neural network method is developed to discriminate direct photons from the neutral pion background in the PHOS spectrometer of the ALICE experiment at the LHC collider. The neural net has been trained to distinguish different classes of events by analyzing the energy-profile tensor of a cluster in its eigen vector coordinate system. Monte-Carlo simulations show that this method diminishes by an order of magnitude the probability of $\pi^0$-meson misidentification as a photon with respect to the direct photon identification efficiency in the energy range up to 120 GeV.