Optimizing many-body atomic descriptors for enhanced computational performance of machine learning based interatomic potentials

Abstract

We explore different ways to simplify the evaluation of the smooth overlap of atomic positions (SOAP) many-body atomic descriptor [Bartók et al., Phys. Rev. B 87, 184115 (2013)]. Our aim is to improve the computational efficiency of SOAP-based similarity kernel construction. While these improved atomic descriptors can be used for general characterization and interpolation of atomic properties, their main target application is accelerated evaluation of machine-learning-based interatomic potentials within the Gaussian approximation potential (GAP) framework [Bartók et al., Phys. Rev. Lett. 104, 136403 (2010)]. We achieve this objective by expressing the atomic densities in an approximate separable form, which decouples the radial and angular channels. We then express the elements of the SOAP descriptor (i.e., the expansion coefficients for the atomic densities) in analytical form given a particular choice of radial basis set. Finally, we derive recursion formulas for the expansion coefficients. This new SOAP-based descriptor allows for tenfold speedups compared to previous implementations, while improving the stability of the radial expansion for distant atomic neighbors, without degradation of the interpolation power of GAP models.

References in corpus (10)

• Practical Bayesian Optimization of Machine Learning Algorithms
• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• On representing chemical environments
• DScribe: Library of Descriptors for Machine Learning in Materials Science
• Comparing molecules and solids across structural and alchemical space
• Machine-learning based interatomic potential for amorphous carbon
• Unified Representation of Molecules and Crystals for Machine Learning
• Atom-Density Representations for Machine Learning
• Growth Mechanism and Origin of High $s{p}^3$ Content in Tetrahedral Amorphous Carbon
• Feature Optimization for Atomistic Machine Learning Yields A Data-Driven Construction of the Periodic Table of the Elements

Cited by in corpus (40)

• Neuroevolution machine learning potentials: Combining high accuracy and low cost in atomistic simulations and application to heat transport
• Physics-inspired structural representations for molecules and materials
• GPUMD: A package for constructing accurate machine-learned potentials and performing highly efficient atomistic simulations
• FCHL revisited: faster and more accurate quantum machine learning
• Representations of molecules and materials for interpolation of quantum-mechanical simulations via machine learning
• Machine-learning interatomic potential for radiation damage and defects in tungsten
• Molecular dynamics simulations of heat transport using machine-learned potentials: A mini review and tutorial on GPUMD with neuroevolution potentials
• Machine Learning for Predicting Thermal Transport Properties of Solids
• Machine learning driven simulated deposition of carbon films: from low-density to diamondlike amorphous carbon
• Quantum-corrected thickness-dependent thermal conductivity in amorphous silicon predicted by machine-learning molecular dynamics simulations
• Global optimization of atomistic structure enhanced by machine learning
• Accurate computational prediction of core-electron binding energies in carbon-based materials: A machine-learning model combining density-functional theory and $\mathbf{GW}$
• The Evolution of Machine Learning Potentials for Molecules, Reactions and Materials
• Ranking the information content of distance measures
• Machine learning force fields based on local parametrization of dispersion interactions: Application to the phase diagram of C${}_{60}$
• Efficient implementation of atom-density representations
• Drawing Phase Diagrams of Random Quantum Systems by Deep Learning the Wave Functions
• Combining phonon accuracy with high transferability in Gaussian approximation potential models
• Tensor-reduced atomic density representations
• Efficiency, Accuracy, and Transferability of Machine Learning Potentials: Application to Dislocations and Cracks in Iron
• Advances in modeling complex materials: The rise of neuroevolution potentials
• A Physically-informed Graph-based Order Parameter for the Universal Characterization of Atomic Structures
• Variable thermal transport in black, blue, and violet phosphorene from extensive atomistic simulations with a neuroevolution potential
• Searching for iron nanoparticles with a general-purpose Gaussian approximation potential
• A general-purpose machine learning Pt interatomic potential for an accurate description of bulk, surfaces and nanoparticles
• Optimal radial basis for density-based atomic representations
• Quantum chemical roots of machine-learning molecular similarity descriptors
• Machine Learning for compositional disorder: A Comparison Between Different Descriptors and Machine Learning Frameworks
• Transferable empirical pseudopotenials from machine learning
• Comment on "Manifolds of quasi-constant SOAP and ACSF fingerprints and the resulting failure to machine learn four body interactions"
• Machine learning at the atomic-scale
• Cluster-based multidimensional scaling embedding tool for data visualization
• Machine learning based modeling of disordered elemental semiconductors: understanding the atomic structure of a-Si and a-C
• EOSnet: Embedded Overlap Structures for Graph Neural Networks in Predicting Material Properties
• Particle Swarm Based Hyper-Parameter Optimization for Machine Learned Interatomic Potentials
• Two-dimensional RMSD projections for reaction path visualization and validation
• Interpretable machine learned predictions of adsorption energies at the metal--oxide interface
• Towards Improved Quantum Machine Learning for Molecular Force Fields
• Improved capabilities of the TurboGAP code for radiation induced cascade simulations: an illustration with silicon
• Compressing local atomic neighbourhood descriptors