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

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

doi:10.1088/1748-0221/13/08/P08023

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

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

Research output: Contribution to journal › Article › peer-review

35 Scopus citations

Abstract

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.

Original language	English
Article number	P08023
Journal	Journal of Instrumentation
Volume	13
Issue number	8
DOIs	https://doi.org/10.1088/1748-0221/13/08/P08023
State	Published - Aug 29 2018

Bibliographical note

Publisher Copyright:
© 2018 IOP Publishing Ltd and Sissa Medialab.

Funding

We thank Adam Coates (Baidu) for helpful and encouraging discussions and Evan Racah (NERSC, LBNL) for support of deep learning applications at NERSC at the beginning of this work. We thank the Erlangen Regional Computing Center (RRZE) for the compute resources and support. We thank the Deutsche Forschungsgemeinschaft (DFG) for the support of this study. EXO-200 data analysis and simulation uses resources of the National Energy Research Scientific Computing Center (NERSC), which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. We gratefully acknowledge the support of Nvidia Corporation with the donation of the Titan Xp GPU used for this research. We thank Adam Coates (Baidu) for helpful and encouraging discussions and Evan Racah (NERSC, LBNL) for support of deep learning applications at NERSC at the beginning of this work. We thank the Erlangen Regional Computing Center (RRZE) for the compute resources and support. We thank the Deutsche Forschungsgemeinschaft (DFG) for the support of this study. EXO-200 data analysis and simulation uses resources of the National Energy Research Scientific Computing Center (NERSC), which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. EXO-200 is supported by DOE and NSF in the U.S., NSERC in Canada, SNF in Switzerland, IBS in Korea, RFBR in Russia, DFG in Germany, and CAS and ISTCP in China. We gratefully acknowledge the KARMEN Collaboration for supplying the cosmic-ray veto detectors, and the WIPP for their hospitality.

Funders	Funder number
U.S. Department of Energy Chinese Academy of Sciences Guangzhou Municipal Science and Technology Project Oak Ridge National Laboratory Extreme Science and Engineering Discovery Environment National Science Foundation National Energy Research Scientific Computing Center National Natural Science Foundation of China	1654495
U.S. Department of Energy Chinese Academy of Sciences Guangzhou Municipal Science and Technology Project Oak Ridge National Laboratory Extreme Science and Engineering Discovery Environment National Science Foundation National Energy Research Scientific Computing Center National Natural Science Foundation of China
U.S. Department of Energy Oak Ridge National Laboratory U.S. Department of Energy National Science Foundation National Energy Research Scientific Computing Center	DE-AC02-05CH11231
U.S. Department of Energy Oak Ridge National Laboratory U.S. Department of Energy National Science Foundation National Energy Research Scientific Computing Center
National Science Foundation Office of International Science and Engineering
Lawrence Berkeley National Laboratory
Natural Sciences and Engineering Research Council of Canada
Deutsche Forschungsgemeinschaft	EXO-200
Deutsche Forschungsgemeinschaft
Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung
Russian Foundation for Basic Research
Chinese Academy of Sciences
International Science and Technology Cooperation Programme

Keywords

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

ASJC Scopus subject areas

Instrumentation
Mathematical Physics

Access to Document

10.1088/1748-0221/13/08/P08023

Cite this

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., ... Zeldovich, O. Y. (2018). Deep neural networks for energy and position reconstruction in EXO-200. Journal of Instrumentation, 13(8), Article P08023. https://doi.org/10.1088/1748-0221/13/08/P08023

@article{3f5b787e87a6479a83d4c6a44fe062ea,

title = "Deep neural networks for energy and position reconstruction in EXO-200",

abstract = "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.",

keywords = "Analysis and statistical methods, Double-beta decay detectors, Pattern recognition, cluster finding, calibration and fitting methods, Time projection chambers",

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

note = "Publisher Copyright: {\textcopyright} 2018 IOP Publishing Ltd and Sissa Medialab.",

year = "2018",

month = aug,

day = "29",

doi = "10.1088/1748-0221/13/08/P08023",

language = "English",

volume = "13",

number = "8",

}

Delaquis, S, Jewell, MJ, Ostrovskiy, I, Weber, M, Ziegler, T, Dalmasson, J, Kaufman, LJ, Richards, T, Albert, JB, Anton, G, Badhrees, I, Barbeau, PS, Bayerlein, R, Beck, D, Belov, V, Breidenbach, M, Brunner, T, Cao, GF, Cen, WR, Chambers, C, Cleveland, B, Coon, M, Craycraft, A, Cree, W, Daniels, T, Danilov, M, Daugherty, SJ, Daughhetee, J, Davis, J, Mesrobian-Kabakian, AD, Devoe, R, Dilling, J, Dolgolenko, A, Dolinski, MJ, Fairbank, W, Farine, J, Feyzbakhsh, S, Fierlinger, P, Fudenberg, D, Gornea, R, Gratta, G, Hall, C, Hansen, EV, 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, KS, Lan, Y, Leonard, DS, Li, GS, Li, S, Licciardi, C, Lin, YH, Maclellan, R, Michel, T, Mong, B, Moore, D, Murray, K, Njoya, O, Odian, A, Piepke, A, Pocar, A, Retière, F, Robinson, AL, Rowson, PC, Schmidt, S, Schubert, A, Sinclair, D, Soma, AK, Stekhanov, V, Tarka, M, Todd, J, Tolba, T, Veeraraghavan, V, Vuilleumier, JL, Wagenpfeil, M, Waite, A, Watkins, J, Wen, LJ, Wichoski, U, Wrede, G, Xia, Q, Yang, L, Yen, YR & Zeldovich, OY 2018, 'Deep neural networks for energy and position reconstruction in EXO-200', Journal of Instrumentation, vol. 13, no. 8, P08023. https://doi.org/10.1088/1748-0221/13/08/P08023

TY - JOUR

T1 - Deep neural networks for energy and position reconstruction in EXO-200

AU - Delaquis, S.

AU - Jewell, M. J.

AU - Ostrovskiy, I.

AU - Weber, M.

AU - Ziegler, T.

AU - Dalmasson, J.

AU - Kaufman, L. J.

AU - Richards, T.

AU - Albert, J. B.

AU - Anton, G.

AU - Badhrees, I.

AU - Barbeau, P. S.

AU - Bayerlein, R.

AU - Beck, D.

AU - Belov, V.

AU - Breidenbach, M.

AU - Brunner, T.

AU - Cao, G. F.

AU - Cen, W. R.

AU - Chambers, C.

AU - Cleveland, B.

AU - Coon, M.

AU - Craycraft, A.

AU - Cree, W.

AU - Daniels, T.

AU - Danilov, M.

AU - Daugherty, S. J.

AU - Daughhetee, J.

AU - Davis, J.

AU - Mesrobian-Kabakian, A. Der

AU - Devoe, R.

AU - Dilling, J.

AU - Dolgolenko, A.

AU - Dolinski, M. J.

AU - Fairbank, W.

AU - Farine, J.

AU - Feyzbakhsh, S.

AU - Fierlinger, P.

AU - Fudenberg, D.

AU - Gornea, R.

AU - Gratta, G.

AU - Hall, C.

AU - Hansen, E. V.

AU - Harris, D.

AU - Hoessl, J.

AU - Hufschmidt, P.

AU - Hughes, M.

AU - Iverson, A.

AU - Jamil, A.

AU - Johnson, A.

AU - Karelin, A.

AU - Koffas, T.

AU - Kravitz, S.

AU - Krücken, R.

AU - Kuchenkov, A.

AU - Kumar, K. S.

AU - Lan, Y.

AU - Leonard, D. S.

AU - Li, G. S.

AU - Li, S.

AU - Licciardi, C.

AU - Lin, Y. H.

AU - Maclellan, R.

AU - Michel, T.

AU - Mong, B.

AU - Moore, D.

AU - Murray, K.

AU - Njoya, O.

AU - Odian, A.

AU - Piepke, A.

AU - Pocar, A.

AU - Retière, F.

AU - Robinson, A. L.

AU - Rowson, P. C.

AU - Schmidt, S.

AU - Schubert, A.

AU - Sinclair, D.

AU - Soma, A. K.

AU - Stekhanov, V.

AU - Tarka, M.

AU - Todd, J.

AU - Tolba, T.

AU - Veeraraghavan, V.

AU - Vuilleumier, J. L.

AU - Wagenpfeil, M.

AU - Waite, A.

AU - Watkins, J.

AU - Wen, L. J.

AU - Wichoski, U.

AU - Wrede, G.

AU - Xia, Q.

AU - Yang, L.

AU - Yen, Y. R.

AU - Zeldovich, O. Ya

PY - 2018/8/29

Y1 - 2018/8/29

N2 - 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.

AB - 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.

KW - Analysis and statistical methods

KW - Double-beta decay detectors

KW - Pattern recognition, cluster finding, calibration and fitting methods

KW - Time projection chambers

UR - http://www.scopus.com/inward/record.url?scp=85053125001&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85053125001&partnerID=8YFLogxK

U2 - 10.1088/1748-0221/13/08/P08023

DO - 10.1088/1748-0221/13/08/P08023

M3 - Article

AN - SCOPUS:85053125001

SN - 1748-0221

VL - 13

JO - Journal of Instrumentation

JF - Journal of Instrumentation

IS - 8

M1 - P08023

ER -

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

Abstract

Bibliographical note

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this