Bonn 2020 – wissenschaftliches Programm

Die DPG-Frühjahrstagung in Bonn musste abgesagt werden! Lesen Sie mehr ...

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

T: Fachverband Teilchenphysik

T 91: Machine Learning: Event and jet reconstruction

T 91.7: Vortrag

Freitag, 3. April 2020, 12:30–12:45, H-HS I

Particle identification with the Belle II Calorimeter using Machine Learning — •Abtin Narimani Charan and Torben Ferber — Deutsches Elektronen-Synchrotron (DESY)

The Belle II experiment, located at the asymmetric SuperKEKB e⁺ e⁻ collider in Tsukuba, Japan, plans to perform studies of B-physics and searches for new physics at the luminosity frontier. The Belle II electromagnetic calorimeter 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. In particular for lower momentum muons and pions which do not reach the outer muon detector, the electromagnetic calorimeter can be critical for muon vs. pion separation. This is crucial for the study of semi-tauonic and semi-leptonic B decays.

This talk presents an application of a convolutional neural network in order to tackle this challenge. Such a network uses the granularity of the calorimeter crystals to provide 5×5 and 7×7 images of calorimeter clusters that contain information of the spatial location of the crystals’ energy deposits from extrapolated tracks. The cluster images of muons and pions are distinguishable since pions undergo hadronic shower in addition to ionization, making the deposited energy more dispersed. In this talk, the performance of the network is investigated with MC samples of muons and pions selected from the Belle II simulation together with data samples which were collected in 2019. Moreover, comparisons will be presented benchmarking against independent approaches to calorimeter-based particle identification.