002746483 001__ 2746483
002746483 005__ 20250108040511.0
002746483 0248_ $$aoai:cds.cern.ch:2746483$$pcerncds:FULLTEXT$$pcerncds:CERN:FULLTEXT$$pcerncds:CERN
002746483 0247_ $$2DOI$$9Frontiers$$a10.3389/frai.2021.649917
002746483 037__ $$9arXiv$$aarXiv:2012.01301$$cphysics.data-an
002746483 037__ $$aFERMILAB-PUB-20-641-ND
002746483 035__ $$9arXiv$$aoai:arXiv.org:2012.01301
002746483 035__ $$9Inspire$$aoai:inspirehep.net:1834492$$d2025-01-07T15:07:55Z$$h2025-01-08T03:00:10Z$$mmarcxml$$ttrue$$uhttps://inspirehep.net/api/oai2d
002746483 035__ $$9Inspire$$a1834492
002746483 041__ $$aeng
002746483 100__ $$aAcciarri, R.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 245__ $$9Frontiers$$aCosmic Ray Background Removal With Deep Neural Networks in SBND
002746483 246__ $$9arXiv$$aCosmic Background Removal with Deep Neural Networks in SBND
002746483 269__ $$c2020-12-02
002746483 260__ $$c2021-08-24
002746483 300__ $$a17 p
002746483 520__ $$9Frontiers$$aIn liquid argon time projection chambers exposed to neutrino beams and running on or near surface levels, cosmic muons and other cosmic particles are incident on the detectors while a single neutrino-induced event is being recorded. In practice, this means that data from surface liquid argon time projection chambers will be dominated by cosmic particles, both as a source of event triggers and as the majority of the particle count in true neutrino-triggered events. In this work, we demonstrate a novel application of deep learning techniques to remove these background particles by applying semantic segmentation on full detector images from the SBND detector, the near detector in the Fermilab Short-Baseline Neutrino Program. We use this technique to identify, at single image-pixel level, whether recorded activity originated from cosmic particles or neutrino interactions.
002746483 520__ $$9arXiv$$aIn liquid argon time projection chambers exposed to neutrino beams and running on or near surface levels, cosmic muons and other cosmic particles are incident on the detectors while a single neutrino-induced event is being recorded. In practice, this means that data from surface liquid argon time projection chambers will be dominated by cosmic particles, both as a source of event triggers and as the majority of the particle count in true neutrino-triggered events. In this work, we demonstrate a novel application of deep learning techniques to remove these background particles by applying semantic segmentation on full detector images from the SBND detector, the near detector in the Fermilab Short-Baseline Neutrino Program. We use this technique to identify, at single image-pixel level, whether recorded activity originated from cosmic particles or neutrino interactions.
002746483 540__ $$3preprint$$aarXiv nonexclusive-distrib 1.0$$uhttp://arxiv.org/licenses/nonexclusive-distrib/1.0/
002746483 540__ $$3publication$$aCC-BY-4.0$$bFrontiers$$uhttp://creativecommons.org/licenses/by/4.0/
002746483 542__ $$3publication$$dThe authors$$g2021
002746483 65017 $$2arXiv$$aphysics.data-an
002746483 65017 $$2SzGeCERN$$aOther Fields of Physics
002746483 690C_ $$aCERN
002746483 690C_ $$aARTICLE
002746483 700__ $$aAdams, C.$$uArgonne$$vArgonne National Laboratory, United States
002746483 700__ $$aAndreopoulos, C.$$uLiverpool U.$$uRutherford$$vUniversity of Liverpool, United Kingdom$$vSTFC, Rutherford Appleton Laboratory, United Kingdom
002746483 700__ $$aAsaadi, J.$$uTexas U., Arlington$$vUniversity of Texas at Arlington, United States
002746483 700__ $$aBabicz, M.$$uCERN$$vCERN, European Organization for Nuclear Research, Switzerland
002746483 700__ $$aBackhouse, C.$$uUniversity Coll. London$$vUniversity College London, United Kingdom
002746483 700__ $$aBadgett, W.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aBagby, L.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aBarker, D.$$uSheffield U.$$vDepartment of Physics and Astronomy, University of Sheffield, United Kingdom
002746483 700__ $$aBasque, V.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aBazetto, M.C. Q.$$uCampinas State U.$$vUniversidade Estadual de Campinas, Brazil$$vCenter for Information Technology Renato Archer Campinas, Brazil
002746483 700__ $$aBetancourt, M.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aBhanderi, A.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aBhat, A.$$uSyracuse U.$$vSyracuse University, United States
002746483 700__ $$aBonifazi, C.$$uUFRJ, Rio de Janeiro$$vUniversidade Federal do Rio de Janeiro, Brazil
002746483 700__ $$aBrailsford, D.$$uLancaster U.$$vLancaster University, United Kingdom
002746483 700__ $$aBrandt, A.G.$$uTexas U., Arlington$$vUniversity of Texas at Arlington, United States
002746483 700__ $$aBrooks, T.$$uSheffield U.$$vDepartment of Physics and Astronomy, University of Sheffield, United Kingdom
002746483 700__ $$aCarneiro, M.F.$$uBrookhaven Natl. Lab.$$vBrookhaven National Laboratory, United States
002746483 700__ $$aChen, Y.$$uBern U.$$vUniversität Bern, Switzerland
002746483 700__ $$aChen, H.$$uBrookhaven Natl. Lab.$$vBrookhaven National Laboratory, United States
002746483 700__ $$aChisnall, G.$$uSussex U.$$vUniversity of Sussex, United Kingdom
002746483 700__ $$aCrespo-Anadón, J.I.$$uMadrid, CIEMAT$$vCIEMAT, Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas, Spain
002746483 700__ $$aCristaldo, E.$$uAsuncion Natl. U.$$vFIUNA Facultad de Ingeniería, Universidad Nacional de Asunción, Paraguay
002746483 700__ $$aCuesta, C.$$uMadrid, CIEMAT$$vCIEMAT, Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas, Spain
002746483 700__ $$ade Icaza Astiz, I.L.$$uSussex U.$$vUniversity of Sussex, United Kingdom
002746483 700__ $$aDe Roeck, A.$$uCERN$$vCERN, European Organization for Nuclear Research, Switzerland
002746483 700__ $$ade Sa Pereira, G.$$uLiverpool U.$$uRutherford$$vUniversity of Liverpool, United Kingdom$$vSTFC, Rutherford Appleton Laboratory, United Kingdom
002746483 700__ $$aDel Tutto, M.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aDi Benedetto, V.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aEreditato, A.$$uBern U.$$vUniversität Bern, Switzerland
002746483 700__ $$aEvans, J.J.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aEzeribe, A.C.$$uSheffield U.$$vDepartment of Physics and Astronomy, University of Sheffield, United Kingdom
002746483 700__ $$aFitzpatrick, R.S.$$uMichigan U.$$vUniversity of Michigan, United States
002746483 700__ $$aFleming, B.T.$$uYale U.$$vWright Laboratory, Department of Physics, Yale University, United States
002746483 700__ $$aForeman, W.$$uIIT, Chicago$$vIllinois Institute of Technology, United States
002746483 700__ $$aFranco, D.$$uYale U.$$vWright Laboratory, Department of Physics, Yale University, United States
002746483 700__ $$aFuric, I.$$uFlorida U.$$vUniversity of Florida, United States
002746483 700__ $$aFurmanski, A.P.$$uMinnesota U.$$vUniversity of Minnesota, United States
002746483 700__ $$aGao, S.$$uBrookhaven Natl. Lab.$$vBrookhaven National Laboratory, United States
002746483 700__ $$aGarcia-Gamez, D.$$uCAFPE, Granada$$vUniversidad de Granada, Spain
002746483 700__ $$aFrandini, H.$$uCampinas State U.$$vUniversidade Estadual de Campinas, Brazil
002746483 700__ $$aGe, G.$$uColumbia U.$$vColumbia University, United States
002746483 700__ $$aGil-Botella, I.$$uMadrid, CIEMAT$$vCIEMAT, Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas, Spain
002746483 700__ $$aGollapinni, S.$$uLos Alamos$$vLos Alamos National Laboratory, United States$$vUniversity of Tennessee, United States
002746483 700__ $$aGoodwin, O.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aGreen, P.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aGriffith, W.C.$$uSussex U.$$vUniversity of Sussex, United Kingdom
002746483 700__ $$aGuenette, R.$$uHarvard U.$$vHarvard University, United States
002746483 700__ $$aGuzowski, P.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aHam, T.$$uLiverpool U.$$vUniversity of Liverpool, United Kingdom
002746483 700__ $$aHenzerling, J.$$uLiverpool U.$$vUniversity of Liverpool, United Kingdom
002746483 700__ $$aHolin, A.$$uUniversity Coll. London$$vUniversity College London, United Kingdom
002746483 700__ $$aHoward, B.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aJones, R. S.$$uLiverpool U.$$vUniversity of Liverpool, United Kingdom
002746483 700__ $$aKalra, D.$$uColumbia U.$$vColumbia University, United States
002746483 700__ $$aKaragiorgi, G.$$uColumbia U.$$vColumbia University, United States
002746483 700__ $$aKashur, L.$$uColorado State U.$$vColorado State University, United States
002746483 700__ $$aKetchum, W.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aKim, M.J.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aKudryavtsev, V.A.$$uSheffield U.$$vDepartment of Physics and Astronomy, University of Sheffield, United Kingdom
002746483 700__ $$aLarkin, J.$$uBrookhaven Natl. Lab.$$vBrookhaven National Laboratory, United States
002746483 700__ $$aLay, H.$$uLancaster U.$$vLancaster University, United Kingdom
002746483 700__ $$aLepetic, I.$$uRutgers U., Piscataway$$vRutgers University, United States
002746483 700__ $$aLittlejohn, B.R.$$uIIT, Chicago$$vIllinois Institute of Technology, United States
002746483 700__ $$aLouis, W.C.$$uLos Alamos$$vLos Alamos National Laboratory, United States
002746483 700__ $$aMachado, A. A.$$uU. Campinas$$vUniversidade Estadual de Campinas, Brazil
002746483 700__ $$aMalek, M.$$uSheffield U.$$vDepartment of Physics and Astronomy, University of Sheffield, United Kingdom
002746483 700__ $$aMardsen, D.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aMariani, C.$$uVirginia Tech.$$vCenter for Neutrino Physics, Virginia Tech, United States
002746483 700__ $$aMarinho, F.$$uSao Carlos Federal U.$$vUniversidade Federal de São Carlos, Brazil
002746483 700__ $$aMastbaum, A.$$uRutgers U., Piscataway$$vRutgers University, United States
002746483 700__ $$aMavrokoridis, K.$$uLiverpool U.$$vUniversity of Liverpool, United Kingdom
002746483 700__ $$aMcConkey, N.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aMeddage, V.$$uFlorida U.$$vUniversity of Florida, United States
002746483 700__ $$aMéndez, D.P.$$uBrookhaven Natl. Lab.$$vBrookhaven National Laboratory, United States
002746483 700__ $$aMettler, T.$$uBern U.$$vUniversität Bern, Switzerland
002746483 700__ $$aMistry, K.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aMogan, A.$$uU. Tennessee, Knoxville$$vUniversity of Tennessee, United States
002746483 700__ $$aMolina, J.$$uAsuncion Natl. U.$$vFIUNA Facultad de Ingeniería, Universidad Nacional de Asunción, Paraguay
002746483 700__ $$aMooney, M.$$uColorado State U.$$vColorado State University, United States
002746483 700__ $$aMora, L.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aMoura, C.A.$$uABC Federal U.$$vUniversidade Federal do ABC, Brazil
002746483 700__ $$aMousseau, J.$$uU. Michigan, Ann Arbor$$vUniversity of Michigan, United States
002746483 700__ $$aNavrer-Agasson, A.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aNicolas-Arnaldos, F. J.$$uGranada U.$$vUniversidad de Granada, Spain
002746483 700__ $$aNowak, J.A.$$uLancaster U.$$vLancaster University, United Kingdom
002746483 700__ $$aPalamara, O.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aPandey, V.$$uFlorida U.$$vUniversity of Florida, United States
002746483 700__ $$aPater, J.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aPaulucci, L.$$uABC Federal U.$$vUniversidade Federal do ABC, Brazil
002746483 700__ $$aPimentel, V.L.$$uCampinas State U.$$vUniversidade Estadual de Campinas, Brazil$$vCenter for Information Technology Renato Archer Campinas, Brazil
002746483 700__ $$aPsihas, F.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aPutnam, G.$$uChicago U., EFI$$vEnrico Fermi Institute, University of Chicago, United States
002746483 700__ $$aQian, X.$$uBrookhaven Natl. Lab.$$vBrookhaven National Laboratory, United States
002746483 700__ $$aRaguzin, E.$$uBrookhaven Natl. Lab.$$vBrookhaven National Laboratory, United States
002746483 700__ $$aRay, H.$$uFlorida U.$$vUniversity of Florida, United States
002746483 700__ $$aReggiani-Guzzo, M.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aRivera, D.$$uPennsylvania U.$$vUniversity of Pennsylvania, United States
002746483 700__ $$aRoda, M.$$uLiverpool U.$$vUniversity of Liverpool, United Kingdom
002746483 700__ $$aRoss-Lonergan, M.$$uColumbia U.$$vColumbia University, United States
002746483 700__ $$aScanavini, G.$$uYale U.$$vWright Laboratory, Department of Physics, Yale University, United States
002746483 700__ $$aScarff, A.$$uSheffield U.$$vDepartment of Physics and Astronomy, University of Sheffield, United Kingdom
002746483 700__ $$aSchmitz, D.W.$$uChicago U., EFI$$vEnrico Fermi Institute, University of Chicago, United States
002746483 700__ $$aSchukraft, A.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aSegreto, E.$$uU. Campinas$$vUniversidade Estadual de Campinas, Brazil
002746483 700__ $$aSoares Nunes, M.$$uSyracuse U.$$vSyracuse University, United States
002746483 700__ $$aSoderberg, M.$$uSyracuse U.$$vSyracuse University, United States
002746483 700__ $$aSöldner-Rembold, S.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aSpitz, J.$$uMichigan U.$$vUniversity of Michigan, United States
002746483 700__ $$aSpooner, N.J. C.$$uSheffield U.$$vDepartment of Physics and Astronomy, University of Sheffield, United Kingdom
002746483 700__ $$aStancari, M.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aStenico, G.V.$$uCampinas State U.$$vUniversidade Estadual de Campinas, Brazil
002746483 700__ $$aSzelc, A.$$uManchester U.$$vUniversity of Manchester, United Kingdom
002746483 700__ $$aTang, W.$$uTennessee U.$$vUniversity of Tennessee, United States
002746483 700__ $$aTena Vidal, J.$$uLiverpool U.$$vUniversity of Liverpool, United Kingdom
002746483 700__ $$aTorretta, D.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aToups, M.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aTouramanis, C.$$uLiverpool U.$$vUniversity of Liverpool, United Kingdom
002746483 700__ $$aTripathi, M.$$uFlorida U.$$vUniversity of Florida, United States
002746483 700__ $$aTufanli, S.$$uCERN$$vCERN, European Organization for Nuclear Research, Switzerland
002746483 700__ $$aTyley, E.$$uSheffield U.$$vDepartment of Physics and Astronomy, University of Sheffield, United Kingdom
002746483 700__ $$aValdiviesso, G.A.$$uAlfenas Fed. U., Pocos de Caldas$$vUniversidade Federal de Alfenas, Brazil
002746483 700__ $$aWorcester, E.$$uBrookhaven Natl. Lab.$$vBrookhaven National Laboratory, United States
002746483 700__ $$aWorcester, M.$$uBrookhaven Natl. Lab.$$vBrookhaven National Laboratory, United States
002746483 700__ $$aYarbrough, G.$$uU. Tennessee, Knoxville$$vUniversity of Tennessee, United States
002746483 700__ $$aYu, J.$$uTexas U., Arlington$$vUniversity of Texas at Arlington, United States
002746483 700__ $$aZamorano, B.$$uGranada U.$$vUniversidad de Granada, Spain
002746483 700__ $$aZennamo, J.$$uFermilab$$vFermi National Accelerator Laboratory, United States
002746483 700__ $$aZglam, A.$$uSheffield U.$$vDepartment of Physics and Astronomy, University of Sheffield, United Kingdom
002746483 710__ $$gSBND Collaboration
002746483 773__ $$c649917$$pFront. Artif. Intell.$$v4$$y2021
002746483 8564_ $$uhttps://lss.fnal.gov/archive/2020/pub/fermilab-pub-20-641-nd.pdf$$yFermilab Library Server (fulltext available)
002746483 8564_ $$82349804$$s3596834$$uhttps://cds.cern.ch/record/2746483/files/frai-04-649917.pdf
002746483 8564_ $$82267640$$s66587$$uhttps://cds.cern.ch/record/2746483/files/loss.png$$y00012 The training progression of the baseline model, trained for 25k iterations. The light blue curve is the training performance at each step, overlaid with a smoothed representation of the same data, and a smoothed representation of the test set.
002746483 8564_ $$82267641$$s45703$$uhttps://cds.cern.ch/record/2746483/files/biggerbatch_iou.png$$y00015 Metric performance across neutrino interaction types, as a function of neutrino energy. The solid lines are the Intersection over Union for the neutrino predicted/labeled pixels, while the dashed lines are the Intersection over Union for the cosmic predicted/labeled pixels. Each color in this plot represents the IoU for all events containing that particular neutrino interaction.
002746483 8564_ $$82267642$$s310127$$uhttps://cds.cern.ch/record/2746483/files/network.png$$y00009 A representation of the multi-plane UResNet architecture. Only two of the three planes are shown in this image for clarity.
002746483 8564_ $$82267643$$s291308$$uhttps://cds.cern.ch/record/2746483/files/Slide2.png$$y00004 The raw data for one image in the dataset at full resolution. Charge observed is colored with blue for smaller charge depositions and red for larger charge depositions. There is an electron neutrino charged current interaction in the Upper TPC.
002746483 8564_ $$82267644$$s64926$$uhttps://cds.cern.ch/record/2746483/files/acc_non_bkg.png$$y00011 The training progression of the baseline model, trained for 25k iterations. The light blue curve is the training performance at each step, overlaid with a smoothed representation of the same data, and a smoothed representation of the test set.
002746483 8564_ $$82267645$$s20457$$uhttps://cds.cern.ch/record/2746483/files/energy.png$$y00002 Neutrino energy of interactions produced for this analysis. Most neutral current events are produced by muon-type neutrinos, and so the $\nu_\mu$ CC and Neutral Current energy spectra are similar. The relative populations here are for the dataset used in this paper, while in the neutrino beam the muon neutrino interactions are far more frequent than electron neutrino.
002746483 8564_ $$82267646$$s148337$$uhttps://cds.cern.ch/record/2746483/files/Slide4.png$$y00006 The labels for the images in Figure~\ref{fig:raw_data} in the dataset at full resolution. White pixels are background, gray pixels are associated with cosmic particles, and red pixels are associated with a neutrino interaction. Plane 2 shows a case of overlap between cosmic and neutrino pixels.
002746483 8564_ $$82267647$$s66551$$uhttps://cds.cern.ch/record/2746483/files/cosmic_iou.png$$y00013 The training progression of the baseline model, trained for 25k iterations. The light blue curve is the training performance at each step, overlaid with a smoothed representation of the same data, and a smoothed representation of the test set.
002746483 8564_ $$82267648$$s4742954$$uhttps://cds.cern.ch/record/2746483/files/2012.01301.pdf$$yFulltext
002746483 8564_ $$82267649$$s167931$$uhttps://cds.cern.ch/record/2746483/files/Slide3.png$$y00005 The raw data for one image in the dataset at full resolution. Charge observed is colored with blue for smaller charge depositions and red for larger charge depositions. There is an electron neutrino charged current interaction in the Upper TPC.
002746483 8564_ $$82267650$$s36106$$uhttps://cds.cern.ch/record/2746483/files/PixelOccupancy.png$$y00010 Distribution of pixel occupancies, by label, in this dataset. In general, the cosmic-labeled pixels are less than 1\% of pixels and the neutrino-labeled pixels are less than 0.3\%
002746483 8564_ $$82267651$$s81251$$uhttps://cds.cern.ch/record/2746483/files/neutrino_iou.png$$y00014 The training progression of the baseline model, trained for 25k iterations. The light blue curve is the training performance at each step, overlaid with a smoothed representation of the same data, and a smoothed representation of the test set.
002746483 8564_ $$82267652$$s142909$$uhttps://cds.cern.ch/record/2746483/files/SBND_TPC.png$$y00000 An illustration of the SBND TPC used in this work. In this image, a neutrino interacts in the left TPC, and the outgoing particles cross the central cathode into the right TPC. The top-down projection images (vertical wire planes) are shown, which are combined into one image as seen in Figure~\ref{fig:raw_data}.
002746483 8564_ $$82267653$$s150913$$uhttps://cds.cern.ch/record/2746483/files/Slide6.png$$y00008 The labels for the images in Figure~\ref{fig:raw_data} in the dataset at full resolution. White pixels are background, gray pixels are associated with cosmic particles, and red pixels are associated with a neutrino interaction. Plane 2 shows a case of overlap between cosmic and neutrino pixels.
002746483 8564_ $$82267654$$s2575342$$uhttps://cds.cern.ch/record/2746483/files/sbnd_cryostat_open.png$$y00001 Engineering diagram of the SBND LArTPC and its surrounding subsystems. Here, the TPC is shown lifted above the cryostat for clarity.
002746483 8564_ $$82267655$$s290284$$uhttps://cds.cern.ch/record/2746483/files/Slide1.png$$y00003 The raw data for one image in the dataset at full resolution. Charge observed is colored with blue for smaller charge depositions and red for larger charge depositions. There is an electron neutrino charged current interaction in the Upper TPC.
002746483 8564_ $$82267656$$s155029$$uhttps://cds.cern.ch/record/2746483/files/Slide5.png$$y00007 The labels for the images in Figure~\ref{fig:raw_data} in the dataset at full resolution. White pixels are background, gray pixels are associated with cosmic particles, and red pixels are associated with a neutrino interaction. Plane 2 shows a case of overlap between cosmic and neutrino pixels.
002746483 8564_ $$82278265$$s4398292$$uhttps://cds.cern.ch/record/2746483/files/fermilab-pub-20-641-nd.pdf$$yFulltext
002746483 8564_ $$82349804$$s157011$$uhttps://cds.cern.ch/record/2746483/files/frai-04-649917.jpg?subformat=icon-700$$xicon-700
002746483 8564_ $$82349804$$s16980$$uhttps://cds.cern.ch/record/2746483/files/frai-04-649917.jpg?subformat=icon-180$$xicon-180
002746483 8564_ $$82349804$$s17653$$uhttps://cds.cern.ch/record/2746483/files/frai-04-649917.gif?subformat=icon$$xicon
002746483 8564_ $$82358907$$s3681394$$uhttps://cds.cern.ch/record/2746483/files/frai-04-649917-A.pdf$$yFulltext from Publisher
002746483 8564_ $$82386082$$s155029$$uhttps://cds.cern.ch/record/2746483/files/w7_Slide5.png$$y00007 The labels for the images in Figure~\ref{fig:raw_data} in the dataset at full resolution. White pixels are background, gray pixels are associated with cosmic particles, and red pixels are associated with a neutrino interaction. Plane 2 shows a case of overlap between cosmic and neutrino pixels.
002746483 8564_ $$82386083$$s290284$$uhttps://cds.cern.ch/record/2746483/files/w3_Slide1.png$$y00003 The raw data for one image in the dataset at full resolution. Charge observed is colored with blue for smaller charge depositions and red for larger charge depositions. There is an electron neutrino charged current interaction in the Upper TPC.
002746483 8564_ $$82386084$$s2127257$$uhttps://cds.cern.ch/record/2746483/files/w1_sbnd_cryostat_open.png$$y00001 Engineering diagram of the SBND LArTPC and its surrounding subsystems. Here, the TPC is shown lifted above the cryostat for clarity.
002746483 8564_ $$82386085$$s167931$$uhttps://cds.cern.ch/record/2746483/files/w5_Slide3.png$$y00005 The raw data for one image in the dataset at full resolution. Charge observed is colored with blue for smaller charge depositions and red for larger charge depositions. There is an electron neutrino charged current interaction in the Upper TPC.
002746483 8564_ $$82386086$$s67268$$uhttps://cds.cern.ch/record/2746483/files/w11_acc_non_bkg.png$$y00011 The training progression of the baseline model, trained for 25k iterations. The light blue curve is the training performance at each step, overlaid with a smoothed representation of the same data, and a smoothed representation of the test set.
002746483 8564_ $$82386087$$s68929$$uhttps://cds.cern.ch/record/2746483/files/w12_loss.png$$y00012 The training progression of the baseline model, trained for 25k iterations. The light blue curve is the training performance at each step, overlaid with a smoothed representation of the same data, and a smoothed representation of the test set.
002746483 8564_ $$82386088$$s291308$$uhttps://cds.cern.ch/record/2746483/files/w4_Slide2.png$$y00004 The raw data for one image in the dataset at full resolution. Charge observed is colored with blue for smaller charge depositions and red for larger charge depositions. There is an electron neutrino charged current interaction in the Upper TPC.
002746483 8564_ $$82386089$$s68893$$uhttps://cds.cern.ch/record/2746483/files/w13_cosmic_iou.png$$y00013 The training progression of the baseline model, trained for 25k iterations. The light blue curve is the training performance at each step, overlaid with a smoothed representation of the same data, and a smoothed representation of the test set.
002746483 8564_ $$82386090$$s145251$$uhttps://cds.cern.ch/record/2746483/files/w0_SBND_TPC.png$$y00000 An illustration of the SBND TPC used in this work. In this image, a neutrino interacts in the left TPC, and the outgoing particles cross the central cathode into the right TPC. The top-down projection images (vertical wire planes) are shown, which are combined into one image as seen in Figure~\ref{fig:raw_data}.
002746483 8564_ $$82386091$$s310127$$uhttps://cds.cern.ch/record/2746483/files/w9_network.png$$y00009 A representation of the multi-plane UResNet architecture. Only two of the three planes are shown in this image for clarity.
002746483 8564_ $$82386092$$s83593$$uhttps://cds.cern.ch/record/2746483/files/w14_neutrino_iou.png$$y00014 The training progression of the baseline model, trained for 25k iterations. The light blue curve is the training performance at each step, overlaid with a smoothed representation of the same data, and a smoothed representation of the test set.
002746483 8564_ $$82386093$$s20457$$uhttps://cds.cern.ch/record/2746483/files/w2_energy.png$$y00002 Neutrino energy of interactions produced for this analysis. Most neutral current events are produced by muon-type neutrinos, and so the $\nu_\mu$ CC and Neutral Current energy spectra are similar. The relative populations here are for the dataset used in this paper, while in the neutrino beam the muon neutrino interactions are far more frequent than electron neutrino.
002746483 8564_ $$82386094$$s150913$$uhttps://cds.cern.ch/record/2746483/files/w8_Slide6.png$$y00008 The labels for the images in Figure~\ref{fig:raw_data} in the dataset at full resolution. White pixels are background, gray pixels are associated with cosmic particles, and red pixels are associated with a neutrino interaction. Plane 2 shows a case of overlap between cosmic and neutrino pixels.
002746483 8564_ $$82386095$$s148337$$uhttps://cds.cern.ch/record/2746483/files/w6_Slide4.png$$y00006 The labels for the images in Figure~\ref{fig:raw_data} in the dataset at full resolution. White pixels are background, gray pixels are associated with cosmic particles, and red pixels are associated with a neutrino interaction. Plane 2 shows a case of overlap between cosmic and neutrino pixels.
002746483 8564_ $$82386096$$s36106$$uhttps://cds.cern.ch/record/2746483/files/w10_PixelOccupancy.png$$y00010 Distribution of pixel occupancies, by label, in this dataset. In general, the cosmic-labeled pixels are less than 1\% of pixels and the neutrino-labeled pixels are less than 0.3\%
002746483 960__ $$a13
002746483 980__ $$aARTICLE