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

AKPIK: Arbeitskreis Physik, moderne Informationstechnologie und Künstliche Intelligenz

AKPIK 2: Data Analytics & Machine Learning

AKPIK 2.5: Vortrag

Mittwoch, 23. März 2022, 17:15–17:30, AKPIK-H13

Ephemeral Learning - Augmenting Triggers with online-trained normalizing flows — •Sascha Diefenbacher — Institut für Experimentalphysik, Universität Hamburg, Germany

The high collision rates at the Large Hadron Collider (LHC) make it impossible to store every single observed interaction. For this reason, only a small subset that passes so-called triggers -- which select potentially interesting events -- are saved while the remainder is discarded. This makes it difficult to perform searches in regions that are usually ignored by trigger setups, for example at low energies. However a sufficiently efficient data compression method could help these searches by storing information about more events than can be saved offline. We investigate the use of a generative machine learning model (specifically a normalizing flow) for the purpose of this compression. The model is trained to learn the underlying data structure of collisions events in an online setting, meaning we can never have a repeated look at past data. After the training the underlying distribution encoded into the network parameters can be analyzed and, for example, probed for anomalies. We initially demonstrate this method for a simple bump hunt, showing that the online trained flow model can recover sensitivity compared to a classical trigger setup. We then extend this demonstration to more complex examples using the LHC Olympics Anomaly Detection Challenge dataset.