Anomaly detection in BDX-MINI experiment data analysis

Seminario di Gruppo III INFN/DIFI

Data inizio

03/12/2025 10:00

Data fine

03/12/2025 12:00

Luogo

Dipartimento di Fisica - A602

Relatori

Patrick Moran
College of William&Mary

AI-supported methods demostrated to be superior for detecting anomalies in data. We applied different types of AI-supported analysis to BDX-MINI, showing that a supervides ML technique provides a better upper limit than traditional analysis, while an unsupervised algorithn offers a completerly unbiased prediction.

Ultimo aggiornamento