Heidelberg 2022 – wissenschaftliches Programm

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

T: Fachverband Teilchenphysik

T 69: DAQ and Trigger 3

T 69.6: Vortrag

Mittwoch, 23. März 2022, 17:30–17:45, T-H28

Anomaly detection as a new strategy for the CMS Trigger — •Sven Bollweg, Karim El Morabit, Gregor Kasieczka, and Artur Lobanov — University of Hamburg, Germany

There exist strong hints for the existence of physics beyond the standard model (BSM). To search for such BSM processes at the LHC, potential candidate events first need to be selected. At CMS, the first selection step is the Level 1 (L1) trigger, which decides whether an event is stored for further analysis. The trigger decision is usually based on criteria motivated by specific models. Another strategy uses the idea that BSM events differ from events originating from standard model (SM) processes. A trigger decision could then utilize this difference to detect anomalous event properties.

This talk discusses such an anomaly detection trigger based on neural networks. An autoencoder (AE) network is trained to reproduce SM events. Using the AE to reproduce BSM events with anomalous properties leads to a reduced quality which can be used for the trigger decision. Since the L1 trigger has a very limited time for the decision, the AE needs to be deployed on dedicated hardware in the form of field programmable gate arrays which presents additional challenges.