Demonstration of background rejection using deep convolutional neural networks in the NEXT experiment

• Published in: The journal of high energy physics Vol. 2021; no. 1; pp. 1 - 22
• Main Authors: Kekic, M., Adams, C., Woodruff, K., Renner, J., Church, E., Del Tutto, M., Hernando Morata, J. A., Gómez-Cadenas, J. J., Álvarez, V., Arazi, L., Arnquist, I. J., Azevedo, C. D. R., Bailey, K., Ballester, F., Benlloch-Rodríguez, J. M., Borges, F. I. G. M., Byrnes, N., Cárcel, S., Carrión, J. V., Cebrián, S., Conde, C. A. N., Contreras, T., Díaz, G., Díaz, J., Diesburg, M., Escada, J., Esteve, R., Felkai, R., Fernandes, A. F. M., Fernandes, L. M. P., Ferrario, P., Ferreira, A. L., Freitas, E. D. C., Generowicz, J., Ghosh, S., Goldschmidt, A., González-Díaz, D., Guenette, R., Gutiérrez, R. M., Haefner, J., Hafidi, K., Hauptman, J., Henriques, C. A. O., Herrero, P., Herrero, V., Ifergan, Y., Jones, B. J. P., Labarga, L., Laing, A., Lebrun, P., López-March, N., Losada, M., Mano, R. D. P., Martín-Albo, J., Martínez, A., Martínez-Lema, G., Martínez-Vara, M., McDonald, A. D., Meziani, Z.-E., Monrabal, F., Monteiro, C. M. B., Mora, F. J., Muñoz Vidal, J., Novella, P., Nygren, D. R., Palmeiro, B., Para, A., Pérez, J., Querol, M., Redwine, A. B., Ripoll, L., Rodríguez García, Y., Rodríguez, J., Rogers, L., Romeo, B., Romo-Luque, C., Santos, F. P., dos Santos, J. M. F., Simón, A., Sofka, C., Sorel, M., Stiegler, T., Toledo, J. F., Torrent, J., Usón, A., Veloso, J. F. C. A., Webb, R., Weiss-Babai, R., White, J. T., Yahlali, N.
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.01.2021; Springer Nature B.V; Springer Berlin; SpringerOpen
• Subjects: Artificial neural networks; Beta decay; Classical and Quantum Gravitation; CNN; Computer simulation; Computer vision; Dark Matter and Double Beta Decay (experiments); Elementary Particles; event classification; Gamma rays; High energy physics; high-pressure xenon chambers; INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND TECHNOLOGY; Neural networks; Neutrinoless double beta decay; NEXT experiment; NUCLEAR PHYSICS AND RADIATION PHYSICS; Pair production; Physics; Physics and Astronomy; PHYSICS OF ELEMENTARY PARTICLES AND FIELDS; Quantum Field Theories; Quantum Field Theory; Quantum Physics; Regular Article - Experimental Physics; Rejection; Relativity Theory; String Theory; Topology; Xenon; Dark Matter and Double Beta Decay (experiments)
• ISSN: 1029-8479; 1029-8479
• DOI: 10.1007/JHEP01(2021)189

A bstract 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 136 Xe. 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 228 Th 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.