Deep neural networks for energy and position reconstruction in EXO-200

Journal Article · 29 August 2018 · Journal of Instrumentation

DOI: https://doi.org/10.1088/1748-0221/13/08/p08023 · OSTI ID:1475444

Delaquis, S.; Jewell, M. J.; Ostrovskiy, I.; Weber, M.; Ziegler, T.; Dalmasson, J.; Kaufman, L. J.; Richards, T.; Albert, J. B.; Anton, G.; Badhrees, I.; Barbeau, P. S.; Bayerlein, R.; Beck, D.; Belov, V.; Breidenbach, M.; Brunner, T.; Cao, G. F.; Cen, W. R.; Chambers, C. ; Cleveland, B.; Coon, M.; Craycraft, A.; Cree, W.; Daniels, T.; Danilov, M.; Daugherty, S. J.; Daughhetee, J.; Davis, J.; Mesrobian-Kabakian, A. Der; DeVoe, R.; Dilling, J.; Dolgolenko, A.; Dolinski, M. J.; Jr., W. Fairbank; Farine, J.; Feyzbakhsh, S.; Fierlinger, P.; Fudenberg, D.; Gornea, R.; Gratta, G.; Hall, C.; Hansen, E. V.; Harris, D.; Hoessl, J.; Hufschmidt, P.; Hughes, M.; Iverson, A.; Jamil, A.; Johnson, A.; Karelin, A.; Koffas, T.; Kravitz, S.; Krücken, R.; Kuchenkov, A.; Kumar, K. S.; Lan, Y.; Leonard, D. S.; Li, G. S.; Li, S.; Licciardi, C.; Lin, Y. H.; MacLellan, R.; Michel, T.; Mong, B.; Moore, D.; Murray, K.; Njoya, O.; Odian, A.; Piepke, A.; Pocar, A.; Retière, F.; Robinson, A. L.; Rowson, P. C.; Schmidt, S.; Schubert, A.; Sinclair, D.; Soma, A. K.; Stekhanov, V.; Tarka, M.; Todd, J.; Tolba, T.; Veeraraghavan, V.; Vuilleumier, J. -L.; Wagenpfeil, M.; Waite, A.; Watkins, J.; Wen, L. J.; Wichoski, U.; Wrede, G.; Xia, Q.; Yang, L.; Yen, Y. -R.; Zeldovich, O. Ya.

Here, we apply deep neural networks (DNN) to data from the EXO-200 experiment. In the studied cases, the DNN is able to reconstruct the relevant parameters—total energy and position—directly from raw digitized waveforms, with minimal exceptions. For the first time, the developed algorithms are evaluated on real detector calibration data. The accuracy of reconstruction either reaches or exceeds what was achieved by the conventional approaches developed by EXO-200 over the course of the experiment. Most existing DNN approaches to event reconstruction and classification in particle physics are trained on Monte Carlo simulated events. Such algorithms are inherently limited by the accuracy of the simulation. We describe a unique approach that, in an experiment such as EXO-200, allows to successfully perform certain reconstruction and analysis tasks by training the network on waveforms from experimental data, either reducing or eliminating the reliance on the Monte Carlo.

• Research Organization: SLAC National Accelerator Laboratory (SLAC), Menlo Park, CA (United States)
• Sponsoring Organization: USDOE
• Contributing Organization: EXO-200 Collaboration
• Grant/Contract Number: AC02-76SF00515
• OSTI ID: 1475444
• Journal Information: Journal of Instrumentation, Journal Name: Journal of Instrumentation Journal Issue: 08 Vol. 13; ISSN 1748-0221
• Publisher: Institute of Physics (IOP)Copyright Statement
• Country of Publication: United States
• Language: English