基于稀疏卷积神经网络的 MicroBooNE 事件重建语义分割
仪器与探测器
2021-04-07 v2 高能物理 - 实验
摘要
我们展示了语义分割网络 SparseSSNet 的性能，该网络对 MicroBooNE 数据提供像素级分类。MicroBooNE 实验采用液氩时间投影室研究中微子性质与相互作用。SparseSSNet 是一个子流形稀疏卷积神经网络，是 MicroBooNE 的 出现振荡分析中首个基于机器学习的算法。该网络被训练将像素分类为五类，并重新分类为与当前分析更相关的两类。SparseSSNet 的输出是后续分析步骤的关键输入。该技术在液氩时间投影室数据中首次使用，并且在准确率和计算资源利用上均优于先前使用的卷积神经网络。在测试样本上达到的准确率为 。对于完整中微子相互作用模拟，处理一幅图像的时间约为 0.5 秒，内存使用在 1 GB 量级，从而可利用大多数典型的 CPU 工作机。
引用
@article{arxiv.2012.08513,
title = {Semantic Segmentation with a Sparse Convolutional Neural Network for Event Reconstruction in MicroBooNE},
author = {MicroBooNE collaboration and P. Abratenko and M. Alrashed and R. An and J. Anthony and J. Asaadi and A. Ashkenazi and S. Balasubramanian and B. Baller and C. Barnes and G. Barr and V. Basque and L. Bathe-Peters and O. Benevides Rodrigues and S. Berkman and A. Bhanderi and A. Bhat and M. Bishai and A. Blake and T. Bolton and L. Camilleri and D. Caratelli and I. Caro Terrazas and R. Castillo Fernandez and F. Cavanna and G. Cerati and Y. Chen and E. Church and D. Cianci and J. M. Conrad and M. Convery and L. Cooper-Troendle and J. I. Crespo-Anadon and M. Del Tutto and S. R. Dennis and D. Devitt and R. Diurba and R. Dorrill and K. Duffy and S. Dytman and B. Eberly and A. Ereditato and J. J. Evans and G. A. Fiorentini Aguirre and R. S. Fitzpatrick and B. T. Fleming and N. Foppiani and D. Franco and A. P. Furmanski and D. Garcia-Gamez and S. Gardiner and G. Ge and S. Gollapinni and O. Goodwin and E. Gramellini and P. Green and H. Greenlee and W. Gu and R. Guenette and P. Guzowski and L. Hagaman and E. Hall and P. Hamilton and O. Hen and G. A. Horton-Smith and A. Hourlier and R. Itay and C. James and J. Jan de Vries and X. Ji and L. Jiang and J. H. Jo and R. A. Johnson and Y. J. Jwa and N. Kamp and N. Kaneshige and G. Karagiorgi and W. Ketchum and B. Kirby and M. Kirby and T. Kobilarcik and I. Kreslo and R. LaZur and I. Lepetic and K. Li and Y. Li and B. R. Littlejohn and W. C. Louis and X. Luo and A. Marchionni and C. Mariani and D. Marsden and J. Marshall and J. Martin-Albo and D. A. Martinez Caicedo and K. Mason and A. Mastbaum and N. McConkey and V. Meddage and T. Mettler and K. Miller and J. Mills and K. Mistry and T. Mohayai and A. Mogan and J. Moon and M. Mooney and A. F. Moor and C. D. Moore and L. Mora Lepin and J. Mousseau and M. Murphy and D. Naples and A. Navrer-Agasson and R. K. Neely and P. Nienaber and J. Nowak and O. Palamara and V. Paolone and A. Papadopoulou and V. Papavassiliou and S. F. Pate and A. Paudel and Z. Pavlovic and E. Piasetzky and I. Ponce-Pinto and S. Prince and X. Qian and J. L. Raaf and V. Radeka and A. Rafique and M. Reggiani-Guzzo and L. Ren and L. Rochester and J. Rodriguez Rondon and H. E. Rogers and M. Rosenberg and M. Ross-Lonergan and B. Russell and G. Scanavini and D. W. Schmitz and A. Schukraft and W. Seligman and M. H. Shaevitz and R. Sharankova and J. Sinclair and A. Smith and E. L. Snider and M. Soderberg and S. Soldner-Rembold and S. R. Soleti and P. Spentzouris and J. Spitz and M. Stancari and J. St. John and T. Strauss and K. Sutton and S. Sword-Fehlberg and A. M. Szelc and N. Tagg and W. Tang and K. Terao and C. Thorpe and M. Toups and Y. -T. Tsai and M. A. Uchida and T. Usher and W. Van De Pontseele and B. Viren and M. Weber and H. Wei and Z. Williams and S. Wolbers and T. Wongjirad and M. Wospakrik and W. Wu and E. Yandel and T. Yang and G. Yarbrough and L. E. Yates and G. P. Zeller and J. Zennamo and C. Zhang},
journal= {arXiv preprint arXiv:2012.08513},
year = {2021}
}