您当前的位置: > 详细浏览

基于神经网络模型的原子核基态自旋分布的随机相互作用研究

请选择邀稿期刊:

Random interaction study on angular-momentum distribution of nuclear ground state with neural networks

摘要: 利用神经网络模型学习、模拟随机两体系综(TBRE)下的原子核基态自旋分布,并对学习后的模型输入特征进行了分析。这是核物理中利用神经网络模型进行分类的典型应用。研究发现使用本文采用的单隐藏层神经网络模型,精确地描述每个随机相互作用系综内的样本仍比较困难。但是神经网络模型可以相对较好地描述基态自旋的统计性质。这可能是因为神经网络模型学习到了TBRE中基态自旋分布的经验规律。
Abstract: The neural network model is employed to learn and simulate the ground state spin distribution of the nucleus within a stochastic two-system ensemble (TBRE), while analyzing the input characteristics of the learned model. This represents a typical application of NN models for classification in nuclear physics. It's still challenging to use the neural network with only a single hidden layer to accurately describe each sample in the TBRE. However, the neural network model effectively captures the statistical properties of the ground state spin, potentially due to its ability to learn empirical rule governing spin distributions in TBRE.

版本历史

[V1] 2023-11-09 15:40:10 ChinaXiv:202311.00130V1 下载全文
点击下载全文
预览
许可声明
metrics指标
  •  点击量109
  •  下载量11
评论
分享