Structural analysis based on unsupervised learning: Search for a characteristic low-dimensional space by local structures in atomistic simulations
Structural analysis based on unsupervised learning: Search for a characteristic low-dimensional space by local structures in atomistic simulations
复制标题
DOI：
10.1103/physrevb.105.075107
复制
发表时间：
2022-02-03
影响因子：
3.7
通讯作者：
Miyazaki，Tsuyoshi
中科院分区：
文献类型：
--
作者：
Tamura，Ryo;Matsuda，Momo;Miyazaki，Tsuyoshi
由于计算技术的进步和计算能力的提高，材料的原子模拟可以以更高的精度模拟大型系统。在这种最先进的原子模拟中可以观察到复杂的现象。然而，越来越难以理解实际发生的事情和机制，例如，在分子动力学（MD）模拟中。我们提出了一种无监督的机器学习方法来分析目标原子周围的局部结构。所提出的方法，它使用两步局部保持投影（TS-LPP），可以找到一个低维空间，其中每个原子或原子组的数据点的分布可以适当地捕获。我们表明，该方法是有效的，从局部结构的角度分析的结晶，液体和非晶态和熔体淬火过程的MD模拟。所提出的方法被证明在硅单组分系统，硅锗二元系统，和铜单组分系统。
Owing to the advances in computational techniques and the increase in computational power, atomistic simulations of materials can simulate large systems with higher accuracy. Complex phenomena can be observed in such state-of-the-art atomistic simulations. However, it has become increasingly difficult to understand what is actually happening and mechanisms, for example, in molecular dynamics (MD) simulations. We propose an unsupervised machine learning method to analyze the local structure around a target atom. The proposed method, which uses the two-step locality preserving projections (TS-LPP), can find a low-dimensional space wherein the distributions of data points for each atom or groups of atoms can be properly captured. We demonstrate that the method is effective for analyzing the MD simulations of crystalline, liquid, and amorphous states and the melt-quench process from the perspective of local structures. The proposed method is demonstrated on a silicon single-component system, a silicon-germanium binary system, and a copper single-component system.