Identifying structural changes with unsupervised machine learning methods

Journal Article · 06 November 2018 · Physical Review E

DOI: https://doi.org/10.1103/physreve.98.053305 · OSTI ID:1565775

Walker, Nicholas ^[1]; Tam, Ka-Ming ^[1]; Novak, Brian ^[1]; Jarrell, M. ^[1]

---

1. Louisiana State Univ., Baton Rouge, LA (United States)

In this work, unsupervised machine learning methods are used to identify structural changes using the melting point transition in classical molecular dynamics simulations as an example application of the approach. Dimensionality reduction and clustering methods are applied to instantaneous radial distributions of atomic configurations from classical molecular dynamics simulations of metallic systems over a large temperature range. Principal component analysis is used to dramatically reduce the dimensionality of the feature space across the samples using an orthogonal linear transformation that preserves the statistical variance of the data under the condition that the new feature space is linearly independent. From there, k-means clustering is used to partition the samples into solid and liquid phases through a criterion motivated by the geometry of the reduced feature space of the samples, allowing for an estimation of the melting point transition. This pattern criterion is conceptually similar to how humans interpret the data but with far greater throughput, as the shapes of the radial distributions are different for each phase and easily distinguishable by humans. The transition temperature estimates derived from this machine learning approach produce comparable results to other methods on similarly small system sizes. These results show that machine learning approaches can be applied to structural changes in physical systems.

• Research Organization: UT-Battelle LLC/ORNL, Oak Ridge, TN (Unted States); Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States). Oak Ridge Leadership Computing Facility (OLCF)
• Sponsoring Organization: National Science Foundation (NSF); USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21)
• Grant/Contract Number: AC05-00OR22725
• OSTI ID: 1565775
• Report Number(s): arXiv:1802.10127v3
• Journal Information: Physical Review E, Journal Name: Physical Review E Journal Issue: 5 Vol. 98; ISSN PLEEE8; ISSN 2470-0045
• Publisher: American Physical Society (APS)Copyright Statement
• Country of Publication: United States
• Language: English