arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2108.07614cond-mat.mes-hall
收藏

固体的电子响应与深度学习

Electronic Response Quantities of Solids and Deep Learning

Kevin Ryczko, Olivier Malenfant-Thuot, Michel Côté, Isaac Tamblyn

摘要:

我们引入了一个深度神经网络(DNN)框架,称为\textbf{r} eal-space \textbf{a} tomic \textbf{d} composition \textbf{net} work(\textsc{radnet}),它能够对固体进行准确的极化和静态介电函数预测。我们使用这些预测计算玻恩有效电荷,纵向光学横向光学(LO-TO)分裂频率,和拉曼张量的两个原型的例子:GaAs和BN。然后,我们计算的拉曼光谱,并找到良好的协议与\textit{ab initio}技术。\textsc{radnet}与当前的方法一样好,甚至更好。最后,我们讨论了如何\textsc{radnet}规模更大的系统,铺平了道路,预测响应函数的中尺度结构与\textit{ab initio}的准确性。

英文摘要:

We introduce a deep neural network (DNN) framework called the \textbf{r}eal-space \textbf{a}tomic \textbf{d}ecomposition \textbf{net}work (\textsc{radnet}), which is capable of making accurate polarization and static dielectric function predictions for solids. We use these predictions to calculate Born-effective charges, longitudinal optical transverse optical (LO-TO) splitting frequencies, and Raman tensors for two prototypical examples: GaAs and BN. We then compute the Raman spectra, and find excellent agreement with \textit{ab initio} techniques. \textsc{radnet} is as good or better than current methodologies. Lastly, we discuss how \textsc{radnet} scales to larger systems, paving the way for predictions of response functions on meso-scale structures with \textit{ab initio} accuracy.

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。

↑