Exploring a potential energy surface by machine learning for characterizing atomic transport
Exploring a potential energy surface by machine learning for characterizing atomic transport
复制标题
DOI：
10.1103/physrevb.97.125124
复制
发表时间：
2017-10
影响因子：
3.7
通讯作者：
Kenta Kanamori;K. Toyoura;Junya Honda;Kazuki Hattori;Atsuto Seko;Masayuki Karasuyama;Kazuki Shitara;M. Shiga;A. Kuwabara;I. Takeuchi
中科院分区：
文献类型：
--
作者：
Kenta Kanamori;K. Toyoura;Junya Honda;Kazuki Hattori;Atsuto Seko;Masayuki Karasuyama;Kazuki Shitara;M. Shiga;A. Kuwabara;I. Takeuchi
我们提出了一种基于优先选择原子输运优势点的计算原子输运势垒的机器学习方法。该方法从势能面的概率高斯过程模型中产生大量的整个势能面的随机样本，从而能够定义主势点的可能性。在十几个质子在氧化物中扩散的模型算例上，与传统的微动弹性带方法进行了比较，证明了该方法的稳健性和有效性。
We propose a machine-learning method for evaluating the potential barrier governing atomic transport based on the preferential selection of dominant points for the atomic transport. The proposed method generates numerous random samples of the entire potential energy surface (PES) from a probabilistic Gaussian process model of the PES, which enables defining the likelihood of the dominant points. The robustness and efficiency of the method are demonstrated on a dozen model cases for proton diffusion in oxides, in comparison with a conventional nudge elastic band method.