Unsupervised learning-based structural analysis: Search for a characteristic low-dimensional space by local structures in atomistic simulations

Abstract

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 datapoints 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.

arXiv comment16 pages, 13 figures

References in corpus (26)

• The SIESTA method for ab initio order-N materials simulation
• Efficient index handling of multidimensional periodic boundary conditions
• Accurate determination of crystal structures based on averaged local bond order parameters
• Machine Learning Unifies the Modelling of Materials and Molecules
• Active learning of linearly parametrized interatomic potentials
• On-the-fly machine learning force field generation: Application to melting points
• Phase transitions of hybrid perovskites simulated by machine-learning force fields trained on-the-fly with Bayesian inference
• O(N) methods in electronic structure calculations
• Machine learning a general purpose interatomic potential for silicon
• Hierarchical structures of amorphous solids characterized by persistent homology
• High-quality continuous random networks
• The elastic constants of MgSiO3 perovskite at pressures and temperatures of the Earth's mantle
• Recent progress with large-scale ab initio calculations: the CONQUEST code
• Entropy based fingerprint for local crystalline order
• Accurate Force Field for Molybdenum by Machine Learning Large Materials Data
• Autonomously revealing hidden local structures in supercooled liquids
• Liquid-liquid transition in supercooled silicon determined by first-principles simulation
• Large scale and linear scaling DFT with the CONQUEST code
• Growing Correlation Length on Cooling Below the Onset of Caging in a Simulated Glass-Forming Liquid
• Hidden Structural Order in Orthorhombic Tantalum Pentoxide (Ta${}_2$O${}_5$)
• Unsupervised learning for local structure detection in colloidal systems
• Machine Learning for Atomic Forces in a Crystalline Solid: Transferability to Various Temperatures
• In operando active learning of interatomic interaction during large-scale simulations
• Size and Temperature Transferability of Direct and Local Deep Neural Networks for Atomic Forces
• Canonical-ensemble extended Lagrangian Born-Oppenheimer molecular dynamics for the linear scaling density functional theory
• Machine learning forces trained by Gaussian process in liquid states: Transferability to temperature and pressure

Cited by in corpus (1)

• Unsupervised classification of disordered patterns in an oppositely charged colloidal system