TY - JOUR
T1 - Development of a Deep Neural Network for the data analysis of the NeuLAND neutron detector
AU - RB Collaboration
AU - Douma, C. A.
AU - Hoemann, E.
AU - Kalantar-Nayestanaki, N.
AU - Mayer, J.
N1 - Funding Information:
We would like to thank C. A. Marocico, J. Messchendorp and L. R. Zwerver for their helpful discussions about the use of Machine Learning Networks. We also would like to thank K. Boretsky, I. Gašparić, A. M. Heinz, H. T. Johansson and O. Tengblad for the helpful discussions about the results presented in this work and for their useful comments on the manuscript. J. Mayer is funded by the BMBF ( 05P19PKFNA ) and the GSI ( KZILGE1416 ). C. A. Douma is funded by the European Union’s Horizon 2020 research and innovation programme collaboration , see Fig. 14 .
Funding Information:
We would like to thank C. A. Marocico, J. Messchendorp and L. R. Zwerver for their helpful discussions about the use of Machine Learning Networks. We also would like to thank K. Boretsky, I. Ga?pari?, A. M. Heinz, H. T. Johansson and O. Tengblad for the helpful discussions about the results presented in this work and for their useful comments on the manuscript. J. Mayer is funded by the BMBF (05P19PKFNA) and the GSI (KZILGE1416). C. A. Douma is funded by the European Union's Horizon 2020 research and innovation programme collaboration, see Fig. 14.
Publisher Copyright:
© 2020 Elsevier B.V.
PY - 2021/2/21
Y1 - 2021/2/21
N2 - A new Machine Learning algorithm for shower-head identification in the NeuLAND neutron detector is presented. The new algorithm uses densely-connected Deep Neural Networks (DNNs) to properly classify events and clusters, which allows accurate reconstruction of the 4-momenta of the detected neutrons. As data-events recorded with NeuLAND vary quite a lot in size, and not all emitted neutrons always produce signals in the detector, careful pre- and post-processing of the data turned out to be required for letting the DNNs be successful in their classifications. However, after properly implementing these procedures, the new algorithm offers a better efficiency than previously-used algorithms in virtually all investigated scenarios. However, the newly-developed algorithm (as well as previous ones) suffers from systematic uncertainties. These uncertainties mainly arise from the physics lists used in the Geant4 simulations to train the DNNs. They are particularly large for the neutron energy range around 200 MeV and for NeuLAND configurations of few double-planes (slimmed down version of the detector). The accuracy improves with a larger number of double-planes. Furthermore, both model improvements and accurate benchmarks are needed for the currently used Geant4 physics lists to reduce the systematic uncertainties of the new algorithm for high-precision studies. Further improvement of the present DNN algorithm is also needed, especially for experiments that require high precision in the neutron scattering angle reconstruction. However, it seems unlikely that this improvement can be realized using only NeuLAND data.
AB - A new Machine Learning algorithm for shower-head identification in the NeuLAND neutron detector is presented. The new algorithm uses densely-connected Deep Neural Networks (DNNs) to properly classify events and clusters, which allows accurate reconstruction of the 4-momenta of the detected neutrons. As data-events recorded with NeuLAND vary quite a lot in size, and not all emitted neutrons always produce signals in the detector, careful pre- and post-processing of the data turned out to be required for letting the DNNs be successful in their classifications. However, after properly implementing these procedures, the new algorithm offers a better efficiency than previously-used algorithms in virtually all investigated scenarios. However, the newly-developed algorithm (as well as previous ones) suffers from systematic uncertainties. These uncertainties mainly arise from the physics lists used in the Geant4 simulations to train the DNNs. They are particularly large for the neutron energy range around 200 MeV and for NeuLAND configurations of few double-planes (slimmed down version of the detector). The accuracy improves with a larger number of double-planes. Furthermore, both model improvements and accurate benchmarks are needed for the currently used Geant4 physics lists to reduce the systematic uncertainties of the new algorithm for high-precision studies. Further improvement of the present DNN algorithm is also needed, especially for experiments that require high precision in the neutron scattering angle reconstruction. However, it seems unlikely that this improvement can be realized using only NeuLAND data.
KW - Machine Learning
KW - NeuLAND
KW - Neural Networks
KW - Neutron detection
KW - RB
UR - https://www.scopus.com/pages/publications/85098475381
U2 - 10.1016/j.nima.2020.164951
DO - 10.1016/j.nima.2020.164951
M3 - Article
AN - SCOPUS:85098475381
SN - 0168-9002
VL - 990
JO - Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
JF - Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
M1 - 164951
ER -