Spectral neighbor analysis method for automated generation of quantum-accurate interatomic potentials

2014/09/12 by Aidan P. Thompson, Laura P. Swiler, Laura Swiler +4 · 5 citations
Materials Science · Mathematics · Physics and Astronomy · #Advanced Chemical Physics Studies #Artificial intelligence #Atom (system on chip) #Bispectrum #Computer science #Harmonics #Interatomic potential #Machine Learning in Materials Science #Mathematics #Molecular dynamics #Physics #Quantum mechanics #Spectral density #Spectroscopy and Quantum Chemical Studies #Spherical harmonics #Statistical physics #Statistics #Voltage #cond-mat.mtrl-sci #k-nearest neighbors algorithm

paper · pdf · doi:10.1016/j.jcp.2014.12.018

arxiv created 2014/09/12 · openalex publication_date 2014/12/16 · arxiv updated 2015/05/20 · openalex created_date 2016/06/24 · openalex updated_date 2026/08/05

Abstract

We present a new interatomic potential for solids and liquids called Spectral Neighbor Analysis Potential (SNAP). The SNAP potential has a very general form and uses machine-learning techniques to reproduce the energies, forces, and stress tensors of a large set of small configurations of atoms, which are obtained using high-accuracy quantum electronic structure (QM) calculations. The local environment of each atom is characterized by a set of bispectrum components of the local neighbor density projected on to a basis of hyperspherical harmonics in four dimensions. The bispectrum components are the same bond-orientational order parameters employed by the GAP potential [arXiv:0910.1019]. The SNAP potential, unlike GAP, assumes a linear relationship between atom energy and bispectrum components. The linear SNAP coefficients are determined using weighted least-squares linear regression against the full QM training set. This allows the SNAP potential to be fit in a robust, automated manner to large QM data sets using many bispectrum coefficients.

Citations

• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• From ultrasoft pseudopotentials to the projector augmented-wave method
• Generalized Gradient Approximation Made Simple
• Fast Parallel Algorithms for Short-Range Molecular Dynamics
• On representing chemical environments
• Special points for Brillouin-zone integrations
• Efficient iterative schemes forab initiototal-energy calculations using a plane-wave basis set
• Special points for Brillouin-zone integrations
• Quantum Theory of Angular Momentum
• Gaussian Approximation Potential: an interatomic potential derived from first principles Quantum Mechanics

Cited by

• Scalable Data-Driven Basis Selection for Linear Machine Learning Interatomic Potentials
• Scaling Neural-Network-Based Molecular Dynamics with Long-Range Electrostatic Interactions to 51 Nanoseconds per Day
• Evolutionary optimization of PAW data-sets for accurate high pressure simulations
• Bayesian force fields from active learning for simulation of inter-dimensional transformation of stanene
• Harnessing Machine Learning for Quantum-Accurate Predictions of Non-Equilibrium Behavior in 2D Materials
• Neural network potential from bispectrum components: A case study on crystalline silicon