Testing a Neural Network for Anomaly Detection in the CMS Global Trigger Test Crate during Run 3

TL;DR

This paper details the deployment and testing of a neural network autoencoder for anomaly detection in the CMS Level-1 Global Trigger test system during LHC Run 3, enabling real-time validation without disrupting data collection.

Contribution

It introduces the integration of a neural network autoencoder into the CMS trigger system for unbiased anomaly detection during live data taking.

Findings

Successful deployment of the neural network in the test crate

Effective detection of potential new physics signatures

Validation of the algorithm during proton collisions

Abstract

We present the deployment and testing of an autoencoder trained for unbiased detection of new physics signatures in the CMS Level-1 Global Trigger (GT) test crate during LHC Run 3. The GT test crate is a copy of the main GT system, receiving the same input data, but whose output is not used to trigger the readout of CMS, providing a platform for thorough testing of new trigger algorithms on live data, but without interrupting data taking. We describe the integration of the Neural Network into the GT test crate, and the monitoring, testing, and validation of the algorithm during proton collisions.