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 56: Experimental methods III
T 56.5: Vortrag
Mittwoch, 1. April 2020, 17:30–17:45, H-ÜR 1
Bias mitigation in Selective Background Monte Carlo Simulation at Belle II — •Yannick Bross1, Thomas Kuhr1, Andreas Lindner1, and James Kahn2 for the Belle II collaboration — 1Ludwig-Maximilians-Universität München — 2Karlsruhe Institute of Technology
The Belle II experiment is expected to collect a total of 50 ab−1 in its lifetime. Considering the focus of measurements on rare processes with a small branching fraction, a strong statistical understanding of the background is required. However, a significant portion of the simulated data is discarded trivially in the first stage of analysis, warranting a better method of simulation to keep up with the amount of data. For this purpose a neural network is implemented to select the relevant data after the Monte Carlo event generation and then only simulate selected events. Existing methods have shown great success with graph neural networks, but lack a good understanding and control of any biases they introduce during selection. Therefore carefully controlling any biases that arise is an important task to improve the network. In this work we introduce methods of checking and further mitigating potential biases.