arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2008.03151hep-phhep-exnucl-th
收藏

用于电子-质子散射的基于机器学习的事件generator

Machine learning-based event generator for electron-proton scattering

Y. Alanazi, P. Ambrozewicz, M. Battaglieri, A. N. Hiller Blin, M. P. Kuchera, Y. Li, T. Liu, R. E. McClellan, W. Melnitchouk, E. Pritchard, M. Robertson, N. Sat… 展开作者

Y. Alanazi, P. Ambrozewicz, M. Battaglieri, A. N. Hiller Blin, M. P. Kuchera, Y. Li, T. Liu, R. E. McClellan, W. Melnitchouk, E. Pritchard, M. Robertson, N. Sato, R. Strauss, L. Velasco

备注：
20 pages, 8 figures, revised version, modified title, expanded author list
阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

摘要：

我们提出了一种新的基于机器学习的蒙特卡罗事件发生器，使用生成对抗网络（GANs），可以用校准的检测器模拟进行训练，以构建一个顶点级事件发生器，而无需对飞秒尺度物理学进行理论假设。我们的框架包括一个基于GAN的检测器折叠作为一个快速代理模型，模仿检测器模拟器。该框架进行了测试和验证的模拟包容性深非弹性散射数据以及现有的参数化探测器模拟，不确定性量化的基础上的统计自举技术。我们的研究结果首次提供了一个现实的概念验证，以减轻理论偏见，推断重建物理观测所需的顶点级事件分布。

英文摘要：

We present a new machine learning-based Monte Carlo event generator using generative adversarial networks (GANs) that can be trained with calibrated detector simulations to construct a vertex-level event generator free of theoretical assumptions about femtometer scale physics. Our framework includes a GAN-based detector folding as a fast-surrogate model that mimics detector simulators. The framework is tested and validated on simulated inclusive deep-inelastic scattering data along with existing parametrizations for detector simulation, with uncertainty quantification based on a statistical bootstrapping technique. Our results provide for the first time a realistic proof-of-concept to mitigate theory bias in inferring vertex-level event distributions needed to reconstruct physical observables.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

↑