Title: Creating Simple, Interpretable Anomaly Detectors for Jet Substructure
Speaker: Spencer Chang (University of Oregon)
Abstract: Anomaly detection with autoencoders is a popular method to search for new physics in a model-agnostic manner. In this talk, we focus on their application to signals with jet substructure and try to understand these "black boxes" by designing mimickers with a small number of energy flow polynomials as inputs. These mimickers perform comparably to the autoencoder when ordering background events, but nontrivially also match the anomaly detection capabilities of the autoencoder across a variety of signal events. Thus, this approach allows one to create simple, interpretable anomaly detectors.