000002889 001__ 2889
000002889 005__ 20220308092025.0
000002889 037__ $$aBELLE2-TALK-CONF-2022-021
000002889 041__ $$aeng
000002889 100__ $$aAbtin Narimani Charan
000002889 245__ $$aParticle identification with the Belle II calorimeter using machine learning
000002889 260__ $$aDPG Spring Meeting$$c2021-03-18
000002889 300__ $$a14
000002889 520__ $$aThe Belle II experiment, located at the asymmetric SuperKEKB e+e- collider in Tsukuba, Japan, performs studies of B-physics and searches for new physics at the luminosity frontier. The Belle II electromagnetic calorimeter is constructed from 8736 CsI(Tl) scintillator crystals. It is designed to measure the energy deposited by charged and neutral particles. The electromagnetic calorimeter also provides important contributions to the Belle II particle identification system. Identification of low-momentum muons and pions is crucial in the electromagnetic calorimeter if they do not reach the outer muon detector. This talk presents an application of a convolutional neural network to separate muons and pions. The granularity of the calorimeter crystals provides 5x5 and 7x7 pixel images of calorimeter clusters which are used as inputs to the neural network. The performance of the network is investigated with data control samples of muons and pions. Finally, comparisons of the neural network approach with conventional methods and with a BDT using shower-shapes are presented.
000002889 6531_ $$aCNN, PID, ECL, Muon-pion separation
000002889 8560_ $$fabtin.narimani.charan@desy.de
000002889 8564_ $$uhttps://docs.belle2.org/record/2889/files/BELLE2-TALK-CONF-2022-021.pdf