Göttingen 2025 – wissenschaftliches Programm

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

T: Fachverband Teilchenphysik

T 54: Data, AI, Computing, Electronics V (Anomaly Detection, Event Selection)

T 54.2: Vortrag

Mittwoch, 2. April 2025, 16:30–16:45, VG 2.101

Anomaly Detection Using Machine Learning at Belle II — •David Giesegh, Nikolai Krug, and Thomas Kuhr — LMU Munich, Germany

In modern High Energy Physics, searches for New Physics are often inspired by specific theoretical models suggesting extensions to the Standard Model. Since, as of yet, none of these could be experimentally verified, the question arises if we are looking in the wrong places. For this reason recent years have seen increasing interest in model-agnostic alternatives to classical analyses, among them Machine Learning assisted methods such as Anomaly Detection. In this project we explored the application of two specific Anomaly Detection procedures based on autoencoders and density estimation at the Belle II Experiment. It could be shown on simulated data scenarios that both methods have the potential to increase the visibility of an unknown small signal on realistic backgrounds, providing a proof of concept for further development of such methods at Belle II.

Keywords: Belle II; Machine Learning; Anomaly Detection; New Physics