TeaNet: Universal neural network interatomic potential inspired by iterative electronic relaxations
TeaNet: Universal neural network interatomic potential inspired by iterative electronic relaxations
复制标题
DOI：
10.1016/j.commatsci.2022.111280
复制
发表时间：
2022-03-09
影响因子：
3.3
通讯作者：
Li, Ju
中科院分区：
文献类型：
--
作者：
Takamoto, So;Izumi, Satoshi;Li, Ju
计算材料科学迫切需要一组任意化学元素的普遍原子间势。图卷积神经网络（GCN）具有丰富的表达能力，但以前主要用于传输标量和向量，而不是秩为>= 2的张量。由于经典的原子间势是由紧密结合的电子弛豫框架激发的，我们想用GCN表示秩>= 2张量信息的迭代传播。在这里，我们提出了一种称为张量嵌入原子网络（TeaNet）的架构，其中角相互作用通过欧几里得张量、向量和标量的结合转化为图卷积。通过应用残差网络（ResNet）架构和循环GCN权值初始化训练，构建了一个更深层（16层）的GCN，其流程类似于迭代电子松弛。我们的训练数据集是通过密度泛函理论计算生成的，主要是化学和结构随机化的配置。我们证明TeaNet可以令人满意地实现涉及元素周期表上前18个元素（H到Ar）的任意结构和反应，包括C-H分子结构，金属，无定形SiO2和水，表现出惊人的良好性能（能量平均绝对误差19 meV/原子）和涉及H到Ar元素的任意化学反应的鲁棒性。
A universal interatomic potential for an arbitrary set of chemical elements is urgently needed in computational materials science. Graph convolution neural network (GCN) has rich expressive power, but previously was mainly employed to transport scalars and vectors, not rank >= 2 tensors. As classic interatomic potentials were inspired by tight-binding electronic relaxation framework, we want to represent this iterative propagation of rank >= 2 tensor information by GCN. Here we propose an architecture called the tensor embedded atom network (TeaNet) where angular interaction is translated into graph convolution through the incorporation of Euclidean tensors, vectors and scalars. By applying the residual network (ResNet) architecture and training with recurrent GCN weights initialization, a much deeper (16 layers) GCN was constructed, whose flow is similar to an iterative electronic relaxation. Our training dataset is generated by density functional theory calculation of mostly chemically and structurally randomized configurations. We demonstrate that arbitrary structures and reactions involving the first 18 elements on the periodic table (H to Ar) can be realized satisfactorily by TeaNet, including C-H molecular structures, metals, amorphous SiO2, and water, showing surprisingly good performance (energy mean absolute error 19 meV/atom) and robustness for arbitrary chemistries involving elements from H to Ar.