Anomaly detection in the search for new physics
Gregor Kasieczka, Hamburg UniversityAbstract:
Given the absence of new physics discovered in direct searches at colliders, new strategies revolving around the use of machine learning to discover anomalous events in data are increasingly proposed and carried out. This talk will discuss how such anomaly detectors can be trained in a data-driven way, what key challenges are, and how they can be overcome.