Performance and Cost Assessment of Machine Learning Interatomic Potentials
University of California San Diego · University of Göttingen · +3 more institutions
Abstract
Machine learning of the quantitative relationship between local environment descriptors and the potential energy surface of a system of atoms has emerged as a new frontier in the development of interatomic potentials (IAPs). Here, we present a comprehensive evaluation of machine learning IAPs (ML-IAPs) based on four local environment descriptors-atom-centered symmetry functions (ACSF), smooth overlap of atomic positions (SOAP), the spectral neighbor analysis potential (SNAP) bispectrum components, and moment tensors-using a diverse data set generated using high-throughput density functional theory (DFT) calculations. The data set comprising bcc (Li, Mo) and fcc (Cu, Ni) metals and diamond group IV…
Citation impact
- FWCI
- 41.78
- Percentile
- 100%
- References
- 113
Authors
11Topics & keywords
- Density functional theory
- Phonon
- Context (archaeology)
- Basis set
- Atom (system on chip)
- Statistical physics
- Computer science
- Artificial intelligence
- Affordable and clean energy
Funding
- NSNational Science FoundationAwards: NA0003525, 1053575, DE-NA0003525
- UDU.S. Department of EnergyAwards: ACI-1053575, DE-NA0003525, NA0003525, -NA0003525
- NENational Energy Research Scientific Computing Center
- DFDeutsche ForschungsgemeinschaftAwards: Be3264/11-2, 329898176, 18-13-00479
- RSRussian Science FoundationAward: 18-13-00479
- NNNational Nuclear Security AdministrationAwards: DE-NA0003525, NA0003525
- UOUniversity of California, San Diego
- OOOffice of Naval ResearchAwards: N00014-16-1-2621, N00014
- SNSandia National LaboratoriesAwards: NA0003525, DE-NA0003525