Nuclear mass based on the multi-task learning neural network method
Nuclear mass based on the multi-task learning neural network method
复制标题
DOI：
10.1007/s41365-022-01031-z
复制
发表时间：
2022-04
影响因子：
2.8
通讯作者：
Xing Ming;Hongfei Zhang;R. Xu;Xiaodong Sun;Yuan Tian-;Zhijie Ge
中科院分区：
文献类型：
--
作者：
Xing Ming;Hongfei Zhang;R. Xu;Xiaodong Sun;Yuan Tian-;Zhijie Ge
采用新设计的多任务学习人工神经网络（MTL-ANN）研究了基于宏观-微观模型的全局核质量。首先，得到最新的原子质量评估AME2020报告的2095个原子核的核结合能，以及液滴模型（LDM）对每个原子核的拟合结果与AME2020数据的偏差。为了补偿LDM中可能忽略的物理特性，在模型中引入了MTL-ANN方法。与单任务学习（STL）方法相比，该网络具有强大的同时学习多核性质的能力，如结合能和单中子、质子分离能。此外，它在降低过拟合风险和实现更好的预测方面非常有效。因此，对于训练和验证数据集以及测试数据集，使用该核质量模型可以获得良好的预测。其中，LDM的全局均方根（RMS）从约2.4 MeV有效地降低到目前的0.2 MeV， RMS也可以达到约0.2 MeV。与STL相比，对于训练集和验证集，结合能提高3-9%，结合能提高20-30%；对于测试集，偏差的减少甚至可以达到30-40%，这明显说明了当前MTL的优势。
The global nuclear mass based on the macroscopic–microscopic model was studied by applying a newly designed multi-task learning artificial neural network (MTL-ANN). First, the reported nuclear binding energies of 2095 nucleireleased in the latest Atomic Mass Evaluation AME2020 and the deviations between the fitting result of the liquid drop model (LDM) and data from AME2020 for each nucleus were obtained. To compensate for the deviations and investigate the possible ignored physics in the LDM, the MTL-ANN method was introduced in the model. Compared to the single-task learning (STL) method, this new network has a powerful ability to simultaneously learn multi-nuclear properties, such as the binding energies and single neutron and proton separation energies. Moreover, it is highly effective in reducing the risk of overfitting and achieving better predictions. Consequently, good predictions can be obtained using this nuclear mass model for both the training and validation datasets and for the testing dataset. In detail, the global root mean square (RMS) of the binding energy is effectively reduced from approximately 2.4 MeV of LDM to the current 0.2 MeV, and the RMS of,can also reach approximately 0.2 MeV. Moreover, compared to STL, for the training and validation sets, 3–9% improvement can be achieved with the binding energy, and 20–30% improvement for,; for the testing sets, the reduction in deviations can even reach 30–40%, which significantly illustrates the advantage of the current MTL.