Skip to main content

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

A preprint version of the article is available at arXiv.


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 136Xe. 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 228Th 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.


  1. NEXT collaboration, The Next White (NEW) Detector, 2018 JINST 13 P12010 [arXiv:1804.02409] [INSPIRE].

  2. NEXT collaboration, Energy calibration of the NEXT-White detector with 1% resolution near Qββ of 136Xe, JHEP 10 (2019) 230 [arXiv:1905.13110] [INSPIRE].

  3. NEXT collaboration, Demonstration of the event identification capabilities of the NEXT-White detector, JHEP 10 (2019) 052 [arXiv:1905.13141] [INSPIRE].

  4. NEXT collaboration, Radiogenic Backgrounds in the NEXT Double Beta Decay Experiment, JHEP 10 (2019) 051 [arXiv:1905.13625] [INSPIRE].

  5. G. Carleo et al., Machine learning and the physical sciences, Rev. Mod. Phys. 91 (2019) 045002 [arXiv:1903.10563] [INSPIRE].

  6. A. Aurisano et al., A Convolutional Neural Network Neutrino Event Classifier, 2016 JINST 11 P09001 [arXiv:1604.01444] [INSPIRE].

  7. MicroBooNE collaboration, Convolutional Neural Networks Applied to Neutrino Events in a Liquid Argon Time Projection Chamber, 2017 JINST 12 P03011 [arXiv:1611.05531] [INSPIRE].

  8. MicroBooNE collaboration, Deep neural network for pixel-level electromagnetic particle identification in the MicroBooNE liquid argon time projection chamber, Phys. Rev. D 99 (2019) 092001 [arXiv:1808.07269] [INSPIRE].

  9. N. Choma et al., Graph Neural Networks for IceCube Signal Classification, in proceedings of the 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA), Orlando, FL, U.S.A., 17–20 December 2018, pp. 386–391 [arXiv:1809.06166] [INSPIRE].

  10. E. Racah et al., Revealing Fundamental Physics from the Daya Bay Neutrino Experiment using Deep Neural Networks, in proceedings of the 2016 15th IEEE International Conference on Machine Learning and Applications (ICMLA), Anaheim, CA, U.S.A., 18–20 December 2016, pp. 892–897 [arXiv:1601.07621] [INSPIRE].

  11. EXO collaboration, Deep Neural Networks for Energy and Position Reconstruction in EXO-200, 2018 JINST 13 P08023 [arXiv:1804.09641] [INSPIRE].

  12. H. Qiao, C. Lu, X. Chen, K. Han, X. Ji and S. Wang, Signal-background discrimination with convolutional neural networks in the PandaX-III experiment using MC simulation, Sci. China Phys. Mech. Astron. 61 (2018) 101007 [arXiv:1802.03489] [INSPIRE].

    ADS  Article  Google Scholar 

  13. P. Ai, D. Wang, G. Huang and X. Sun, Three-dimensional convolutional neural networks for neutrinoless double-beta decay signal/background discrimination in high-pressure gaseous Time Projection Chamber, 2018 JINST 13 P08015 [arXiv:1803.01482] [INSPIRE].

  14. NEXT collaboration, Background rejection in NEXT using deep neural networks, 2017 JINST 12 T01004 [arXiv:1609.06202] [INSPIRE].

  15. NEXT collaboration, Sensitivity of NEXT-100 to Neutrinoless Double Beta Decay, JHEP 05 (2016) 159 [arXiv:1511.09246] [INSPIRE].

  16. D. Nygren, High-pressure xenon gas electroluminescent TPC for 0-ν ββ-decay search, Nucl. Instrum. Meth. A 603 (2009) 337 [INSPIRE].

    ADS  Article  Google Scholar 

  17. NEXT collaboration, Calibration of the NEXT-White detector using 83mKr decays, 2018 JINST 13 P10014 [arXiv:1804.01780] [INSPIRE].

  18. J. Martín-Albo, The NEXT experiment for neutrinoless double beta decay searches, Ph.D. Thesis, University of Valencia, Valencia Spain (2015) [INSPIRE].

  19. GEANT4 collaboration, GEANT4 — a simulation toolkit, Nucl. Instrum. Meth. A 506 (2003) 250 [INSPIRE].

  20. A. Krizhevsky, I. Sutskever and G.E. Hinton, Imagenet classification with deep convolutional neural networks, Commun. ACM 60 (2017) 84.

    Article  Google Scholar 

  21. N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever and R. Salakhutdinov, Dropout: A simple way to prevent neural networks from overfitting, J. Mach. Learn. Res. 15 (2014) 1929.

    MathSciNet  MATH  Google Scholar 

  22. S. Ioffe and C. Szegedy, Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift, arXiv:1502.03167 [INSPIRE].

  23. C. Guo, G. Pleiss, Y. Sun and K.Q. Weinberger, On calibration of modern neural networks, arXiv:1706.04599.

  24. K. He, X. Zhang, S. Ren and J. Sun, Deep Residual Learning for Image Recognition, arXiv:1512.03385 [INSPIRE].

  25. K. He, X. Zhang, S. Ren and J. Sun, Identity mappings in deep residual networks, arXiv:1603.05027.

  26. X. Li, S. Chen, X. Hu and J. Yang, Understanding the Disharmony Between Dropout and Batch Normalization by Variance Shift, in proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, U.S.A., 15–20 June 2019, pp. 2677–2685.

  27. J. Deng, W. Dong, R. Socher, L. Li, K. Li and L. Fei-Fei, ImageNet: A large-scale hierarchical image database, in proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, U.S.A., 20–25 June 2009, pp. 248–255.

  28. B. Graham and L. van der Maaten, Submanifold sparse convolutional networks, arXiv:1706.01307.

  29. L. Dominé and K. Terao, Scalable deep convolutional neural networks for sparse, locally dense liquid argon time projection chamber data, Phys. Rev. D 102 (2020) 012005 [arXiv:1903.05663] [INSPIRE].

  30. C. Shorten and T.M. Khoshgoftaar, A survey on image data augmentation for deep learning, J. Big Data 6 (2019) 60.

    Article  Google Scholar 

  31. G.J. Székely and M.L. Rizzo, Testing for equal distributions in high dimension, InterStat 5 (2004) 1.

    Google Scholar 

  32. G. Székely and M.L. Rizzo, Energy statistics: A class of statistics based on distances, J. Stat. Plann. Infer. 8 (2013) 1249.

    MathSciNet  Article  Google Scholar 

  33. R.A. Fisher, The Design of Experiments, Oliver and Boyd (1935).

  34. NEXT collaboration, Sensitivity of a tonne-scale NEXT detector for neutrinoless double beta decay searches, arXiv:2005.06467 [INSPIRE].

  35. NEXT collaboration, Initial results of NEXT-DEMO, a large-scale prototype of the NEXT-100 experiment, 2013 JINST 8 P04002 [arXiv:1211.4838] [INSPIRE].

  36. NEXT collaboration, Operation and first results of the NEXT-DEMO prototype using a silicon photomultiplier tracking array, 2013 JINST 8 P09011 [arXiv:1306.0471] [INSPIRE].

Download references

Author information

Authors and Affiliations



Corresponding author

Correspondence to M. Kekic.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

ArXiv ePrint: 2009.10783

NEXT Co-spokesperson. (J. J. Gómez-Cadenas)

On leave from Soreq Nuclear Research Center, Yavneh, Israel. (R. Weiss-Babai)

Deceased (J. T. White)

Rights and permissions

Open Access . This article is distributed under the terms of the Creative Commons Attribution License (CC-BY 4.0), which permits any use, distribution and reproduction in any medium, provided the original author(s) and source are credited.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

The NEXT collaboration., Kekic, M., Adams, C. et al. Demonstration of background rejection using deep convolutional neural networks in the NEXT experiment. J. High Energ. Phys. 2021, 189 (2021).

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • DOI:


  • Dark Matter and Double Beta Decay (experiments)