Machine learning-based event generator for electron-proton scattering

Machine learning-based event generator for electron-proton scattering
复制标题

基于机器学习的电子-质子散射事件生成器

DOI:
10.1103/physrevd.106.096002
复制
发表时间:
2022
期刊:
影响因子:
5
通讯作者:
Pritchard, E.
Pritchard, E.
中科院分区:
物理与天体物理2区
文献类型:
--
作者:
Alanazi, Y.;Ambrozewicz, P.;Battaglieri, M.;Hiller Blin, A. N.;Kuchera, M. P.;Li, Y.;Liu, T.;McClellan, R. E.;Melnitchouk, W.;Pritchard, E.

文献摘要

参考文献

相似文献

我们提出了一种新的基于机器学习的蒙特卡罗事件生成器,它使用生成性对抗网络(GANS),可以用校准的探测器模拟来训练,以构造一个顶点级别的事件生成器,而不需要对飞秒尺度物理的理论假设。我们的框架包括一个基于GaN的探测器折叠,作为模拟探测器模拟器的快速代理模型。该框架在模拟的包含深度非弹性的散射数据以及用于探测器模拟的现有参数上进行了测试和验证,并基于统计自举技术进行了不确定性量化。我们的结果首次提供了一个现实的概念证明,以减少在推断重建物理观测所需的顶点级事件分布时的理论偏差。
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.
DOI: 10.1140/epjc/s10052-015-3710-4
发表时间: 2015-12-08
影响因子: 4.4
作者:
Abramowicz, H.;Abt, I.;Zotkin, D. S.
通讯作者: Zotkin, D. S.
DOI: 10.1007/s41781-018-0015-y
发表时间: 2018-05
影响因子: --
作者:
P. Musella;F. Pandolfi
通讯作者: P. Musella;F. Pandolfi