Background rejection in NEXT using deep neural networks

J. Renner, A. Farbin, J. Muñoz Vidal, J. M. Benlloch-Rodríguez, A. Botas, P. Ferrario, J. J. Gómez-Cadenas, V. Álvarez, C. D.R. Azevedo, F. I.G. Borges, S. Cárcel, J. V. Carrión, S. Cebrián, A. Cervera, C. A.N. Conde, J. Díaz, M. Diesburg, R. Esteve, L. M.P. Fernandes, A. L. FerreiraE. D.C. Freitas, A. Goldschmidt, D. González-Díaz, R. M. Gutiérrez, J. Hauptman, C. A.O. Henriques, J. A.Hernando Morata, V. Herrero, B. Jones, L. Labarga, A. Laing, P. Lebrun, I. Liubarsky, N. López-March, D. Lorca, M. Losada, J. Martín-Albo, G. Martínez-Lema, A. Martínez, F. Monrabal, C. M.B. Monteiro, F. J. Mora, L. M. Moutinho, M. Nebot-Guinot, P. Novella, D. Nygren, B. Palmeiro, A. Para, J. Pérez, M. Querol, L. Ripoll, J. Rodríguez, F. P. Santos, J. M.F. Dos Santos, L. Serra, D. Shuman, A. Simón, C. Sofka, M. Sorel, J. F. Toledo, J. Torrent, Z. Tsamalaidze, J. F.C.A. Veloso, J. White, R. Webb, N. Yahlali, H. Yepes-Ramírez

Research output: Contribution to journalArticlepeer-review


We investigate the potential of using deep learning techniques to reject background events in searches for neutrinoless double beta decay with high pressure xenon time projection chambers capable of detailed track reconstruction. The differences in the topological signatures of background and signal events can be learned by deep neural networks via training over many thousands of events. These networks can then be used to classify further events as signal or background, providing an additional background rejection factor at an acceptable loss of efficiency. The networks trained in this study performed better than previous methods developed based on the use of the same topological signatures by a factor of 1.2 to 1.6, and there is potential for further improvement.

Original languageEnglish (US)
Article numberT01004
JournalJournal of Instrumentation
Issue number1
StatePublished - Jan 16 2017


  • Analysis and statistical methods
  • Double-beta decay detectors
  • Pattern recognition
  • Time projection chambers
  • calibration and fitting methods
  • cluster finding

ASJC Scopus subject areas

  • Instrumentation
  • Mathematical Physics


Dive into the research topics of 'Background rejection in NEXT using deep neural networks'. Together they form a unique fingerprint.

Cite this