Efficient nonparametric $n$-body force fields from machine learning
Abstract
We provide a definition and explicit expressions for -body Gaussian Process (GP) kernels which can learn any interatomic interaction occurring in a physical system, up to -body contributions, for any value of . The series is complete, as it can be shown that the "universal approximator" squared exponential kernel can be written as a sum of -body kernels. These recipes enable the choice of optimally efficient force models for each target system, as confirmed by extensive testing on various materials. We furthermore describe how the -body kernels can be "mapped" on equivalent representations that provide database-size-independent predictions and are thus crucially more efficient. We explicitly carry out this mapping procedure for the first non-trivial (3-body) kernel of the series, and show that this reproduces the GP-predicted forces with accuracy while being orders of magnitude faster. These results open the way to using novel force models (here named "M-FFs") that are computationally as fast as their corresponding standard parametrised -body force fields, while retaining the nonparametric character, the ease of training and validation, and the accuracy of the best recently proposed machine learning potentials.
Cite
@article{arxiv.1801.04823,
title = {Efficient nonparametric $n$-body force fields from machine learning},
author = {Aldo Glielmo and Claudio Zeni and Alessandro De Vita},
journal= {arXiv preprint arXiv:1801.04823},
year = {2018}
}
Comments
13 pages, 8 captioned figures