Heidelberg 2022 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
T: Fachverband Teilchenphysik
T 79: Data Analysis, Information Technology and Artificial Intelligence 4
T 79.4: Talk
Wednesday, March 23, 2022, 17:00–17:15, T-H38
Identification of pions and muons with the Belle II calorimeter using convolutional neural network — •Abtin Narimani Charan1 and Torben Ferber2 for the Belle II collaboration — 1Deutsches Elektronen-Synchrotron (DESY), Hamburg, Germany — 2Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany
The Belle II experiment is located at the asymmetric SuperKEKB e+ e− collider in Tsukuba, Japan. The electromagnetic calorimeter (ECL) in Belle II is designed to reconstruct neutral particles. Additionally, the ECL can identify charged particles e.g. electrons, muons and pions. Identification of low-momenta muons and pions in the ECL is crucial if they do not reach the outer muon detector.
This talk presents an application of a convolutional neural network (CNN) to separate muons and pions in the ECL using energy deposition patterns of 7x7 crystal images. Due to the high level of beam background, the performance of the CNN in samples with different beam background levels is studied. Moreover, the impact of adding pulse-shape and timing information in addition to the energy information is investigated and compared with the previous results in different momentum ranges. Finally, the performance of the network is investigated with data control samples of muons and pions.