Machine Learning a General-Purpose Interatomic Potential for Silicon
Машинное обучение универсального межатомного потенциала для кремния
2018-12-14
SCID: 54.1/kb3t9ftn
Discuss with AI
electronic structure calculationsenergy and forcesmachine-learning interatomic potentialmaterial propertiessilicon
Figures from the paper
Abstract (AI)
A machine-learning based approach to computing the energy and forces of silicon atoms circumvents the need for complex electronic structure calculations, offering an efficient method for predicting material properties that can be extended to other materials.
Key Findings
1
A machine-learning interatomic potential is developed to compute silicon atomic energies and forces.
2
The approach avoids complex electronic-structure calculations, enabling more efficient prediction of silicon material properties.
3
The proposed method is designed as a general-purpose potential and may be extendable to other materials.
Research Object
silicon atomic systems
Research Subject
the energy and forces of silicon atoms, enabling prediction of material properties
Publication Details
Publication Date
2018-12-14
Journal
Physical Review X
Publisher
American Physical Society
ISSN
2160-3308
Cited by
471
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai6
Statistical Learning Theory1999
The atomic simulation environment—a Python library for working with atoms2017
On representing chemical environments2013
Quantum chemistry structures and properties of 134 kilo molecules2014
Machine learning in materials informatics: recent applications and prospects2017
Comparing molecules and solids across structural and alchemical space2016