Resonant Anomaly Detection with Multiple Reference Datasets
This work addresses a limitation in anomaly detection for high energy physics by enabling the use of multiple datasets, offering incremental improvements over single-reference methods.
The paper tackles the problem of resonant anomaly detection in high energy physics by generalizing existing methods to use multiple reference datasets, resulting in improved performance with realistic and synthetic data and providing finite-sample guarantees.
An important class of techniques for resonant anomaly detection in high energy physics builds models that can distinguish between reference and target datasets, where only the latter has appreciable signal. Such techniques, including Classification Without Labels (CWoLa) and Simulation Assisted Likelihood-free Anomaly Detection (SALAD) rely on a single reference dataset. They cannot take advantage of commonly-available multiple datasets and thus cannot fully exploit available information. In this work, we propose generalizations of CWoLa and SALAD for settings where multiple reference datasets are available, building on weak supervision techniques. We demonstrate improved performance in a number of settings with realistic and synthetic data. As an added benefit, our generalizations enable us to provide finite-sample guarantees, improving on existing asymptotic analyses.