.png){kind=icon}
Return
Building machine learning force fields for nanoclusters
.png){kind=icon}
OA
AI
DOI：10.1063/1.5024558.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
We assess Gaussian process (GP) regression as a technique to model interatomic forces in metal nanoclusters by analyzing the performance of 2-body, 3-body, and many-body kernel functions on a set of 19-atom Ni cluster structures. We find that 2-body GP kernels fail to provide faithful force estimates, despite succeeding in bulk Ni systems. However, both 3- and many-body kernels predict forces within an similar to 0.1e V/angstrom average error even for small training datasets and achieve high accuracy even on out-of-sample, high temperature structures. While training and testing on the same structure always provide satisfactory accuracy, cross-testing on dissimilar structures leads to higher prediction errors, posing an extrapolation problem. This can be cured using heterogeneous training on databases that contain more than one structure, which results in a good trade-off between versatility and overall accuracy. Starting from a 3-body kernel trained this way, we build an efficient non-parametric 3-body force field that allows accurate prediction of structural properties at finite temperatures, following a newly developed scheme [A. Glielmo et al., Phys. Rev. B 95, 214302 (2017)]. We use this to assess the thermal stability of Ni-19 nanoclusters at a fractional cost of full ab initio calculations. Published by AIP Publishing.
Keywords:
NEURAL-NETWORK POTENTIALS
STRUCTURAL TRANSITIONS
METAL-CLUSTERS
NANOPARTICLES
SIZE
NANOALLOYS
TEMPERATURES
CATALYSIS
ATOMS
DFT
Journal
IF:
3.1
Papers:
7.2W
Citations:
23.2W
Organization
Cited Papers
ShareSaveShareSaveFluxionality of gold nanoparticles investigated by Born-Oppenheimer molecular dynamics.png)
PHYSICAL REVIEW BOAAI
PHYSICAL REVIEW BIF3.7
OAAIShareSaveFree energy surface of two- and three-dimensional transitions of Au 12 nanoclusters obtained by ab initio metadynamicsPHYSICAL REVIEW B
PHYSICAL REVIEW BIF3.7
PREAI
ShareSaveAccurate interatomic force fields via machine learning with covariant kernelsPHYSICAL REVIEW B
OAAI
PHYSICAL REVIEW BIF3.7
OAAIShareSaveShareSaveShareSave
View more