Vertex finding in neutrino-nucleus interaction: a model architecture comparison
Vertex finding in neutrino-nucleus interaction: a model architecture comparison
复制标题
DOI：
10.1088/1748-0221/17/08/t08013
复制
发表时间：
2022-01
影响因子：
1.3
通讯作者：
F. Akbar;A. Ghosh;Steven R. Young;S. Akhter;Z. A. Dar;V. Ansari;M. Ascencio;M. Athar;A. Bodek;J. L. Bonilla;A. Bravar;H. Budd;G. Caceres;T. Cai;M. Carneiro;G. D'iaz;J. Félix;L. Fields;A. Filkins;R. Fine;P.K.Gaura;R. Gran;D. Harris;D. Jena;S. Jena;J. Kleykamp;A. Klustov'a;D. Last;A. Lozano;X. Lu;E. Maher;S. Manly;W. A. Mann;K. McFarland;B. Messerly;J. Miller;O. Moreno;J. Morf'in;J. Nelson;C. Nguyen;A. Olivier;V. Paolone;G. Perdue;K. Plows;M. Ram'irez;D. Ruterbories;H. Su;V. Syrotenko;A. Waldron;B. Yaeggy;L. Zazueta
中科院分区：
文献类型：
--
作者：
F. Akbar;A. Ghosh;Steven R. Young;S. Akhter;Z. A. Dar;V. Ansari;M. Ascencio;M. Athar;A. Bodek;J. L. Bonilla;A. Bravar;H. Budd;G. Caceres;T. Cai;M. Carneiro;G. D'iaz;J. Félix;L. Fields;A. Filkins;R. Fine;P.K.Gaura;R. Gran;D. Harris;D. Jena;S. Jena;J. Kleykamp;A. Klustov'a;D. Last;A. Lozano;X. Lu;E. Maher;S. Manly;W. A. Mann;K. McFarland;B. Messerly;J. Miller;O. Moreno;J. Morf'in;J. Nelson;C. Nguyen;A. Olivier;V. Paolone;G. Perdue;K. Plows;M. Ram'irez;D. Ruterbories;H. Su;V. Syrotenko;A. Waldron;B. Yaeggy;L. Zazueta
我们比较了用于识别MINERvA探测器中中微子相互作用顶点位置的机器学习算法的不同神经网络架构。将手工开发和优化的架构与使用橡树岭国家实验室开发的“深度学习的多节点进化神经网络”（MENNDL）包以自动化方式开发的架构进行比较。虽然领域专家手动调整的网络性能最好，但差异可以忽略不计，自动生成的网络性能也很好。在网络优化中，人类和计算机资源之间总是存在权衡，这项工作表明，假设资源可用，自动优化提供了一种令人信服的方式来节省大量专家时间。
We compare different neural network architectures for machine learning algorithms designed to identify the neutrino interaction vertex position in the MINERvA detector. The architectures developed and optimized by hand are compared with the architectures developed in an automated way using the package “Multi-node Evolutionary Neural Networks for Deep Learning” (MENNDL), developed at Oak Ridge National Laboratory. While the domain-expert hand-tuned network was the best performer, the differences were negligible and the auto-generated networks performed as well. There is always a trade-off between human, and computer resources for network optimization and this work suggests that automated optimization, assuming resources are available, provides a compelling way to save significant expert time.