1. Demonstration of background rejection using deep convolutional neural networks in the NEXT experiment
- Author
-
T. Contreras, Z. E. Meziani, R. Weiss-Babai, I. J. Arnquist, J.V. Carrión, J. Renner, N. López-March, L. Rogers, F.J. Mora, J. Generowicz, T.M. Stiegler, Vicente Herrero, F.P. Santos, G. Martínez-Lema, C.M.B. Monteiro, R.D.P. Mano, L.M.P. Fernandes, J.M. Benlloch-Rodríguez, P. Herrero, C. Adams, N. Byrnes, B. Palmeiro, J.A. Hernando Morata, N. Yahlali, D. González-Díaz, Y. Rodriguez Garcia, J. S. Díaz, M. Martínez-Vara, P. Novella, F.I.G.M. Borges, Romain Esteve, J.F.C.A. Veloso, E.D.C. Freitas, J. Martin-Albo, M. Del Tutto, A. Usón, R. Guenette, F. Monrabal, Roberto Gutiérrez, A.D. McDonald, C.D.R. Azevedo, J.T. White, J.F. Toledo, S. Cárcel, P. Lebrun, A. Martínez, M. Diesburg, E. Church, A. Laing, Kevin Bailey, J. Rodríguez, M. Kekic, A.B. Redwine, C.A.O. Henriques, J. Escada, L. Ripoll, J. Torrent, Lior Arazi, B. J. P. Jones, Víctor H. Alvarez, J. Haefner, B. Romeo, R. Felkai, M. Losada, A. Goldschmidt, J. Hauptman, K. Woodruff, L. Labarga, Y. Ifergan, J.M.F. dos Santos, C. Romo-Luque, Javier Pérez, S. Cebrián, Sudip Ghosh, R. C. Webb, G. Díaz, F. Ballester, Paola Ferrario, D.R. Nygren, A.F.M. Fernandes, M. Querol, C. Sofka, A. Para, M. Sorel, A.L. Ferreira, K. Hafidi, C.A.N. Conde, A. Simón, J.J. Gómez-Cadenas, J. Muñoz Vidal, and UAM. Departamento de Física Teórica
- Subjects
Nuclear and High Energy Physics ,Physics - Instrumentation and Detectors ,Calibration (statistics) ,Computer Science::Neural and Evolutionary Computation ,Nuclear physics ,FOS: Physical sciences ,Topology (electrical circuits) ,01 natural sciences ,Convolutional neural network ,Atomic ,Partícules (Física nuclear) ,High Energy Physics - Experiment ,Interaccions electró-positró ,TECNOLOGIA ELECTRONICA ,High Energy Physics - Experiment (hep-ex) ,Particle and Plasma Physics ,Double beta decay ,0103 physical sciences ,Dark Matter and Double Beta Decay (experiments) ,Nuclear ,Nuclear Matrix ,lcsh:Nuclear and particle physics. Atomic energy. Radioactivity ,010306 general physics ,Electron-positron interactions ,Mathematical Physics ,Particles (Nuclear physics) ,Physics ,Quantum Physics ,010308 nuclear & particles physics ,business.industry ,Event (computing) ,Network on ,SIGNAL (programming language) ,Molecular ,Física ,Pattern recognition ,Detector ,Instrumentation and Detectors (physics.ins-det) ,Beta Decay ,Nuclear & Particles Physics ,Doble desintegració beta ,Identification (information) ,lcsh:QC770-798 ,Física nuclear ,Artificial intelligence ,business - Abstract
[EN] Convolutional neural networks (CNNs) are widely used state-of-the-art computer vision tools that are becoming increasingly popular in high-energy physics. In this paper, we attempt to understand the potential of CNNs for event classification in the NEXT experiment, which will search for neutrinoless double-beta decay in Xe-136. To do so, we demonstrate the usage of CNNs for the identification of electron-positron pair production events, which exhibit a topology similar to that of a neutrinoless double-beta decay event. These events were produced in the NEXT-White high-pressure xenon TPC using 2.6 MeV gamma rays from a Th-228 calibration source. We train a network on Monte Carlo-simulated events and show that, by applying on-the-fly data augmentation, the network can be made robust against differences between simulation and data. The use of CNNs offers significant improvement in signal efficiency and background rejection when compared to previous non-CNN-based analyses, This study used computing resources from Artemisa, co-funded by the European Union through the 2014-2020 FEDER Operative Programme of the Comunitat Valenciana, project DIFEDER/2018/048. This research used resources of the Argonne Leadership Computing Facility, which is a DOE Office of Science User Facility supported under Contract DE-AC02-06CH11357. The NEXT collaboration acknowledges support from the following agencies and institutions: Xunta de Galicia (Centro singularde investigacion de Galicia accreditation 2019-2022), by European Union ERDF, and by the "Maria de Maeztu" Units of Excellence program MDM-2016-0692 and the Spanish Research State Agency"; the European Research Council (ERC) under the Advanced Grant 339787-NEXT; the European Union's Framework Programme for Research and Innovation Horizon 2020 (2014-2020) under the Grant Agreements No. 674896, 690575 and 740055; the Ministerio de Economia y Competitividad and the Ministerio de Ciencia, Innovacion y Universidades of Spain under grants FIS2014-53371-C04, RTI2018-095979, the Severo Ochoa Program grants SEV-20140398 and CEX2018-000867-S; the GVA of Spain under grants PROMETEO/2016/120 and SEJI/2017/011; the Portuguese FCT under project PTDC/FIS-NUC/2525/2014 and under projects UID/FIS/04559/2020 to fund the activities of LIBPhys-UC; the U.S. Department of Energy under contracts number DE-AC02-07CH11359 (Fermi National Accelerator Laboratory), DE-FG02-13ER42020 (Texas A&M) and DE-SC0019223/DE SC0019054 (University of Texas at Arlington); and the University of Texas at Arlington. DGD acknowledges Ramon y Cajal program (Spain) under contract number RYC-2015 18820. JMA acknowledges support from Fundacion Bancaria "la Caixa" (ID 100010434), grant code LCF/BQ/PI19/11690012. We also warmly acknowledge the Laboratori Nazionali del Gran Sasso (LNGS) and the Dark Side collaboration for their help with TPB coating of various parts of the NEXT-White TPC. Finally, we are grateful to the Laboratorio Subterraneo de Canfranc for hosting and supporting the NEXT experiment.
- Published
- 2021