Speaker
Description
The Online Data Quality Monitoring (DQM) of the CMS electromagnetic calorimeter (ECAL) is a crucial tool that allows ECAL experts to quickly identify, localize, and diagnose a broad range of detector issues that would impact on the quality of the data for physics. A real-time autoencoder-based anomaly detection system using semi-supervised machine learning is presented enabling the detection of anomalies in the CMS electromagnetic calorimeter data. A novel method is introduced which maximizes the anomaly detection performance by exploiting the time-dependent evolution of anomalies as well as spatial variations in the detector response.
The autoencoder-based system, which was previously described in ACAT 2022, is able to efficiently detect anomalies, while maintaining a very low false discovery rate. The focus of this presentation will be on the deployment of this system in the online DQM workflow during LHC Run 3 (2022+), the operational experience during this run period, and the ability to detect issues that are missed by the existing cut-based anomaly detection system.
References
https://doi.org/10.1007/s41781-024-00118-z
| Experiment context, if any | CMS experiment |
|---|
