SBND中基于深度神经网络的宇宙射线本底去除

Cosmic Ray Background Removal With Deep Neural Networks in SBND.

作者信息

Acciarri R, Adams C, Andreopoulos C, Asaadi J, Babicz M, Backhouse C, Badgett W, Bagby L, Barker D, Basque V, Bazetto M C Q, Betancourt M, Bhanderi A, Bhat A, Bonifazi C, Brailsford D, Brandt A G, Brooks T, Carneiro M F, Chen Y, Chen H, Chisnall G, Crespo-Anadón J I, Cristaldo E, Cuesta C, de Icaza Astiz I L, De Roeck A, de Sá Pereira G, Del Tutto M, Di Benedetto V, Ereditato A, Evans J J, Ezeribe A C, Fitzpatrick R S, Fleming B T, Foreman W, Franco D, Furic I, Furmanski A P, Gao S, Garcia-Gamez D, Frandini H, Ge G, Gil-Botella I, Gollapinni S, Goodwin O, Green P, Griffith W C, Guenette R, Guzowski P, Ham T, Henzerling J, Holin A, Howard B, Jones R S, Kalra D, Karagiorgi G, Kashur L, Ketchum W, Kim M J, Kudryavtsev V A, Larkin J, Lay H, Lepetic I, Littlejohn B R, Louis W C, Machado A A, Malek M, Mardsen D, Mariani C, Marinho F, Mastbaum A, Mavrokoridis K, McConkey N, Meddage V, Méndez D P, Mettler T, Mistry K, Mogan A, Molina J, Mooney M, Mora L, Moura C A, Mousseau J, Navrer-Agasson A, Nicolas-Arnaldos F J, Nowak J A, Palamara O, Pandey V, Pater J, Paulucci L, Pimentel V L, Psihas F, Putnam G, Qian X, Raguzin E, Ray H, Reggiani-Guzzo M, Rivera D, Roda M, Ross-Lonergan M, Scanavini G, Scarff A, Schmitz D W, Schukraft A, Segreto E, Soares Nunes M, Soderberg M, Söldner-Rembold S, Spitz J, Spooner N J C, Stancari M, Stenico G V, Szelc A, Tang W, Tena Vidal J, Torretta D, Toups M, Touramanis C, Tripathi M, Tufanli S, Tyley E, Valdiviesso G A, Worcester E, Worcester M, Yarbrough G, Yu J, Zamorano B, Zennamo J, Zglam A

机构信息

Fermi National Accelerator Laboratory, Batavia, IL, United States.

Argonne National Laboratory, Lemont, IL, United States.

出版信息

Front Artif Intell. 2021 Aug 24;4:649917. doi: 10.3389/frai.2021.649917. eCollection 2021.

DOI:10.3389/frai.2021.649917

PMID:34505055

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC8421797/

Abstract

In liquid argon time projection chambers exposed to neutrino beams and running on or near surface levels, cosmic muons, and other cosmic particles are incident on the detectors while a single neutrino-induced event is being recorded. In practice, this means that data from surface liquid argon time projection chambers will be dominated by cosmic particles, both as a source of event triggers and as the majority of the particle count in true neutrino-triggered events. In this work, we demonstrate a novel application of deep learning techniques to remove these background particles by applying deep learning on full detector images from the SBND detector, the near detector in the Fermilab Short-Baseline Neutrino Program. We use this technique to identify, on a pixel-by-pixel level, whether recorded activity originated from cosmic particles or neutrino interactions.

摘要

在暴露于中微子束且运行在地表或接近地表水平的液态氩时间投影室中，当记录单个中微子诱发事件时，宇宙μ子和其他宇宙粒子会入射到探测器上。实际上，这意味着来自地表液态氩时间投影室的数据将以宇宙粒子为主，既是事件触发的来源，也是真正中微子触发事件中粒子计数的主体。在这项工作中，我们展示了深度学习技术的一种新应用，即通过对费米实验室短基线中微子计划中的近探测器SBND探测器的全探测器图像应用深度学习来去除这些背景粒子。我们使用这种技术在逐个像素的层面上识别记录的活动是源自宇宙粒子还是中微子相互作用。