Hyperactive learning for data-driven interatomic potentials

Data-driven interatomic potentials have emerged as a powerful tool for approximating ab initio potential energy surfaces. The most time-consuming step in creating these interatomic potentials is typically the generation of a suitable training database. To aid this process hyperactive learning (HAL),...

Full description

Saved in:
Bibliographic Details
Published in:npj computational materials 2023-01, Vol.9 (1), p.168-168, Article 168
Main Authors: van der Oord, Cas, Sachs, Matthias, Kovács, Dávid Péter, Ortner, Christoph, Csányi, Gábor
Format: Article
Language:English
Subjects:
639/301/1034/1035
639/301/1034/1037
Accuracy
Active learning
Characterization and Evaluation of Materials
Chemistry and Materials Science
Computational Intelligence
Configurations
Energy
Learning
Materials Science
Mathematical and Computational Engineering
Mathematical and Computational Physics
Mathematical Modeling and Industrial Mathematics
Melt temperature
Molecular dynamics
Neural networks
Polyethylene glycol
Potential energy
Simulation
Theoretical
Training
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Data-driven interatomic potentials have emerged as a powerful tool for approximating ab initio potential energy surfaces. The most time-consuming step in creating these interatomic potentials is typically the generation of a suitable training database. To aid this process hyperactive learning (HAL), an accelerated active learning scheme, is presented as a method for rapid automated training database assembly. HAL adds a biasing term to a physically motivated sampler (e.g. molecular dynamics) driving atomic structures towards uncertainty in turn generating unseen or valuable training configurations. The proposed HAL framework is used to develop atomic cluster expansion (ACE) interatomic potentials for the AlSi10 alloy and polyethylene glycol (PEG) polymer starting from roughly a dozen initial configurations. The HAL generated ACE potentials are shown to be able to determine macroscopic properties, such as melting temperature and density, with close to experimental accuracy.
ISSN:2057-3960
2057-3960
DOI:10.1038/s41524-023-01104-6