Heidelberg 2022 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
AKPIK: Arbeitskreis Physik, moderne Informationstechnologie und Künstliche Intelligenz
AKPIK 2: Data Analytics & Machine Learning
AKPIK 2.7: Vortrag
Mittwoch, 23. März 2022, 17:45–18:00, AKPIK-H13
Identifying Slow Pions using Support Vector Machines — •Timo Schellhaas1, Jens Sören Lange2, and Stephanie Käs3 — 1II. Physikalisches Institut, JLU Gießen, Germany — 2II. Physikalisches Institut, JLU Gießen, Germany — 3II. Physikalisches Institut, JLU Gießen, Germany
Finding new physics beyond the standard model is of highest interest. Pions with a low transversal momentum (slow pions) are linked to interesting decay scenarios and are therefore studied at the Belle II experiment. However, it is dificult to detect slow pions due to their low momenta: a large amount of them does only reach the Belle II pixeldetector (PXD), but not the outer detectors (e.g. the drift chamber). In order to improve the detection rate it is suggested to use a machine learning model. One possible model is the support vector machine (SVM) algorithm. Therefore a simulated data set is used to train the SVM model with different parameters, including a modified kernel, with the goal of reaching better results than other models.