Accurate Interatomic Force Fields via Machine Learning with Covariant Kernels

Abstract

We present a novel scheme to accurately predict atomic forces as vector quantities, rather than sets of scalar components, by Gaussian Process (GP) Regression. This is based on matrix-valued kernel functions, on which we impose the requirements that the predicted force rotates with the target configuration and is independent of any rotations applied to the configuration database entries. We show that such covariant GP kernels can be obtained by integration over the elements of the rotation group SO(d) for the relevant dimensionality d. Remarkably, in specific cases the integration can be carried out analytically and yields a conservative force field that can be recast into a pair interaction form. Finally, we show that restricting the integration to a summation over the elements of a finite point group relevant to the target system is sufficient to recover an accurate GP. The accuracy of our kernels in predicting quantum-mechanical forces in real materials is investigated by tests on pure and defective Ni, Fe and Si crystalline systems.

References in corpus (5)

• Machine Learning of Accurate Energy-Conserving Molecular Force Fields
• Big Data of Materials Science - Critical Role of the Descriptor
• Machine-learning based interatomic potential for amorphous carbon
• Nearsightedness of Electronic Matter
• Machine learning for many-body physics: The case of the Anderson impurity model

Cited by in corpus (100)

• Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems
• Towards Exact Molecular Dynamics Simulations with Machine-Learned Force Fields
• Active learning of linearly parametrized interatomic potentials
• Big-Data Science in Porous Materials: Materials Genomics and Machine Learning
• Physics-inspired structural representations for molecules and materials
• Machine learning a general purpose interatomic potential for silicon
• Machine learning for electronically excited states of molecules
• Machine-learning interatomic potentials for materials science
• FCHL revisited: faster and more accurate quantum machine learning
• Physically-informed artificial neural networks for atomistic modeling of materials
• Symmetry-Adapted Machine-Learning for Tensorial Properties of Atomistic Systems
• Automatic Selection of Atomic Fingerprints and Reference Configurations for Machine-Learning Potentials
• WACSF - Weighted Atom-Centered Symmetry Functions as Descriptors in Machine Learning Potentials
• Unified Representation of Molecules and Crystals for Machine Learning
• Achieving DFT accuracy with a machine-learning interatomic potential: thermomechanics and defects in bcc ferromagnetic iron
• An Accurate and Transferable Machine Learning Potential for Carbon
• A Transferable Machine-Learning Model of the Electron Density
• sGDML: Constructing Accurate and Data Efficient Molecular Force Fields Using Machine Learning
• Incorporating long-range physics in atomic-scale machine learning
• Extending the Accuracy of the SNAP Interatomic Potential Form
• On the Completeness of Atomic Structure Representations
• Atom-Density Representations for Machine Learning
• Non-covalent interactions across organic and biological subsets of chemical space: Physics-based potentials parametrized from machine learning
• Accurate Force Field for Molybdenum by Machine Learning Large Materials Data
• Representations of molecules and materials for interpolation of quantum-mechanical simulations via machine learning
• Efficient nonparametric $n$-body force fields from machine learning
• Feature Optimization for Atomistic Machine Learning Yields A Data-Driven Construction of the Periodic Table of the Elements
• Machine Learning on Neutron and X-Ray Scattering
• Predicting molecular dipole moments by combining atomic partial charges and atomic dipoles
• Atomic cluster expansion of scalar, vectorial and tensorial properties and including magnetism and charge transfer
• Operators in Machine Learning: Response Properties in Chemical Space
• Molecular Force Fields with Gradient-Domain Machine Learning: Construction and Application to Dynamics of Small Molecules with Coupled Cluster Forces
• OrbNet Denali: A machine learning potential for biological and organic chemistry with semi-empirical cost and DFT accuracy
• Using Gaussian Process Regression to Simulate the Vibrational Raman Spectra of Molecular Crystals
• Error-Controlled Exploration of Chemical Reaction Networks with Gaussian Processes
• Ultra-fast interpretable machine-learning potentials
• Thermodynamics and dielectric response of ${\text{BaTiO}}_3$ by data-driven modeling
• Recursive evaluation and iterative contraction of $N$-body equivariant features
• Data-driven simulation and characterisation of gold nanoparticle melting
• Direct prediction of phonon density of states with Euclidean neural networks
• Machine learning and excited-state molecular dynamics
• Equivariant representations for molecular Hamiltonians and N-center atomic-scale properties
• Universal QM/MM Approaches for General Nanoscale Applications
• PANNA: Properties from Artificial Neural Network Architectures
• Bayesian Force Fields from Active Learning for Simulation of Inter-Dimensional Transformation of Stanene
• Application of machine learning potentials to predict grain boundary properties in fcc elemental metals
• Molecular Force Fields with Gradient-Domain Machine Learning (GDML): Comparison and Synergies with Classical Force Fields
• Challenges for Machine Learning Force Fields in Reproducing Potential Energy Surfaces of Flexible Molecules
• Dynamical Strengthening of Covalent and Non-Covalent Molecular Interactions by Nuclear Quantum Effects at Finite Temperature
• Building machine learning force fields for nanoclusters
• Group-theoretical high-order rotational invariants for structural representations: Application to linearized machine learning interatomic potential
• Efficient implementation of atom-density representations
• Predicting tensorial molecular properties with equivariant machine-learning models
• Fast Neural Network Approach for Direct Covariant Forces Prediction in Complex Multi-Element Extended Systems
• On Machine Learning Force Fields for Metallic Nanoparticles
• Universal Machine Learning Kohn-Sham Hamiltonian for Materials
• Machine learning potentials for multicomponent systems: The Ti-Al binary system
• Development of a general-purpose machine-learning interatomic potential for aluminum by the physically-informed neural network method
• Automated construction of quantum-classical hybrid models
• Self-Parametrizing System-Focused Atomistic Models
• Gaussian process model of 51-dimensional potential energy surface for protonated imidazole dimer
• Hydration free energies from kernel-based machine learning: Compound-database bias
• Development of a physically-informed neural network interatomic potential for tantalum
• A spectral-neighbour representation for vector fields: machine-learning potentials including spin
• Atomistic Mechanism Underlying the Si(111)-(7\times7) Surface Reconstruction Revealed by Artificial Neural-network Potential
• Machine learning the DFT potential energy surface for inorganic halide perovskite CsPbBr${}_3$
• An orbital-based representation for accurate Quantum Machine Learning
• Compact atomic descriptors enable accurate predictions via linear models
• Anharmonic Thermodynamics of Vacancies Using a Neural Network Potential
• Fast and flexible long-range models for atomistic machine learning
• Optimal radial basis for density-based atomic representations
• Linearized machine-learning interatomic potentials for non-magnetic elemental metals: Limitation of pairwise descriptors and trend of predictive power
• 3DReact: Geometric deep learning for chemical reactions
• SPA${}^{\mathrm{H}}$M: the Spectrum of Approximated Hamiltonian Matrices representations
• MGNN: Moment Graph Neural Network for Universal Molecular Potentials
• Size and Temperature Transferability of Direct and Local Deep Neural Networks for Atomic Forces
• Modeling of many-body interactions between elastic spheres through symmetry functions
• Gaussian Process States: A data-driven representation of quantum many-body physics
• A transferable artificial neural network model for atomic forces in nanoparticles
• Machine-learning of atomic-scale properties based on physical principles
• Construction of Machine Learned Force Fields with Quantum Chemical Accuracy: Applications and Chemical Insights
• A Bayesian Inference Framework for Compression and Prediction of Quantum States
• Building nonparametric $n$-body force fields using Gaussian process regression
• Accurate Molecular Dynamics Enabled by Efficient Physically-Constrained Machine Learning Approaches
• Modeling of effective interactions between ligand coated nanoparticles through symmetry functions
• Machine learning at the atomic-scale
• Interpolating many-body wave functions for accelerated molecular dynamics on the near-exact electronic surface
• Neural-networks model for force prediction in multi-principal-element alloys
• Learning intermolecular forces at liquid-vapor interfaces
• A simple approach to rotationally invariant machine learning of avector quantity
• Predictive power of polynomial machine learning potentials for liquid states in 22 elemental systems
• High-performance descriptor for magnetic materials: Accurate discrimination of magnetic structure
• Learning the Electrostatic Response of the Electron Density through a Symmetry-Adapted Vector Field Model
• Ab initio machine learning in chemical compound space
• Machine-Learned Potentials for Solvation Modeling
• Reconstructing Kernel-based Machine Learning Force Fields with Super-linear Convergence
• A general formalism for machine-learning models based on multipolar-spherical harmonics
• Simplifying inverse material design problems for fixed lattices with alchemical chirality
• Surfing multiple conformation-property landscapes via machine learning: Designing magnetic anisotropy
• Systematic global structure search of bismuth-based binary systems under pressure using machine learning potentials