Benchmarking structural evolution methods for training of machine learned interatomic potentials. (21st September 2022)

• Record Type: Journal Article
• Authors: Waters, Michael J; Rondinelli, James M
• Is Part Of: Journal of physics. Volume 34:Number 38(2022)
• Publication Date: 2022-09-21
• Subjects: machine learning -- interatomic potentials -- density functional theory -- zirconia; Condensed matter -- Periodicals; Matière condensée -- Périodiques; Vaste stoffen; Vloeistoffen; Natuurkunde; Electronic journals; Computer network resources; 530.4105
• DOI: 10.1088/1361-648X/ac7f73
• Languages: English
• ISSNs: 0953-8984

Abstract

Abstract: When creating training data for machine-learned interatomic potentials (MLIPs), it is common to create initial structures and evolve them using molecular dynamics (MD) to sample a larger configuration space. We benchmark two other modalities of evolving structures, contour exploration (CE) and dimer-method (DM) searches against MD for their ability to produce diverse and robust density functional theory training data sets for MLIPs. We also discuss the generation of initial structures which are either from known structures or from random structures in detail to further formalize the structure-sourcing processes in the future. The polymorph-rich zirconium-oxygen composition space is used as a rigorous benchmark system for comparing the performance of MLIPs trained on structures generated from these structural evolution methods. Using Behler–Parrinello neural networks as our MLIP models, we find that CE and the DM searches are generally superior to MD in terms of spatial descriptor diversity and statistical accuracy.