Litcius/Paper detail

Autoencoder-Based Anomaly Detection System for Online Data Quality Monitoring of the CMS Electromagnetic Calorimeter

Daniel Abadjiev, T. Adams, P. Adžić, M. Ahmad, C. Amendola, Michael Benjamin Andrews, R. Arcidiacono, S. Argirò, Andrew Askew, E. Auffray, V. Azzolini, D. Bailleux, R. Band, David Barney, P. Barria, N. Bartosik, Chiara Basile, D. Bastos, K. W. Bell, M. Besançon, Riccardo Lo Bianco, C. Biino, V. Blinov, Cecilia Borca, A. Bornheim, R. M. Brown, M. Campana, S. Castells, F. Cavallari, F. Cetorelli, S. Chatterjee, S. Chatterjee, G. Chaudhary, Jianfei Chen, N. Chernyavskaya, H. Chung, M. Cipriani, L. Cokic, C. Cooke, F. Cossio, F. Couderc, D. Cristoforetti, G. Cucciati, L. Cunqueiro Mendez, D. Da Silva Di Calafiori, I. Dafinei, D. J. A. Cockerill, M. Déjardin, D. Del Re, G. Della Ricca, P. Depasse, J. Dervan, Emanuele Di Marco, M. Diemoz, T. Dimova, G. Dissertori, M. Dittmar, A. Dolgopolov, M. Donegà, M. Dordevic, H. El Mamouni, F. Errico, Felipe Espinosa, J. Fauré, J. Fay, J. Fernández Menéndez, F. Ferri, L. Finco, F. Fiori, Ellery Frahm, W. Funk, T. Gadek, Jakub Gajownik, M. Galli, Serguei Ganjour, S. Gascon, A. Ghezzi, Partha Ghose, S. Gninenko, S. Goadhouse, N. Godinović, N. Golubev, P. Govoni, P. Gras, John Hakala, G. Hamel de Monchenault, A. Harilal, N. Härringer, R. Hashmi, H. F. Heath, R. Hirosky, Ka Wa Ho, Xuehui Hou, Q. Ingram, Sh. Jain, Tahir Javaid, C. P. Jessop, R. Jimènez, B. M. Joshi, E. Jourd‘hui

2024Computing and Software for Big Science21 citationsDOIOpen Access PDF

Abstract

Abstract The CMS detector is a general-purpose apparatus that detects high-energy collisions produced at the LHC. Online data quality monitoring of the CMS electromagnetic calorimeter is a vital operational tool that allows detector experts to quickly identify, localize, and diagnose a broad range of detector issues that could affect the quality of physics data. 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 is able to efficiently detect anomalies, while maintaining a very low false discovery rate. The performance of the system is validated with anomalies found in 2018 and 2022 LHC collision data. In addition, the first results from deploying the autoencoder-based system in the CMS online data quality monitoring workflow during the beginning of Run 3 of the LHC are presented, showing its ability to detect issues missed by the existing system.

Topics & Concepts

AutoencoderAnomaly detectionDetectorCalorimeter (particle physics)Large Hadron ColliderComputer scienceAnomaly (physics)Data qualityReal-time computingData miningArtificial intelligencePhysicsParticle physicsDeep learningEngineeringTelecommunicationsMetric (unit)Operations managementCondensed matter physicsParticle physics theoretical and experimental studiesParticle Detector Development and PerformanceComputational Physics and Python Applications