Efficient nonparametric $n$-body force fields from machine learning

Abstract

We provide a definition and explicit expressions for $n$-body Gaussian Process (GP) kernels which can learn any interatomic interaction occurring in a physical system, up to $n$-body contributions, for any value of $n$. The series is complete, as it can be shown that the "universal approximator" squared exponential kernel can be written as a sum of $n$-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 $n$-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 $\text{meV/}\mathring{A}$ 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 $n$-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.

13 pages, 8 captioned figures

References in corpus (8)

• A Spectral Analysis Method for Automated Generation of Quantum-Accurate Interatomic Potentials
• Machine-learning based interatomic potential for amorphous carbon
• Alchemical and structural distribution based representation for improved QML
• Symmetry-Adapted Machine-Learning for Tensorial Properties of Atomistic Systems
• On the accuracy of the MB-pol many-body potential for water: Interaction energies, vibrational frequencies, and classical thermodynamic and dynamical properties from clusters to liquid water and ice
• Understanding molecular representations in machine learning: The role of uniqueness and target similarity
• A Machine Learning Potential for Graphene
• Non-covalent interactions across organic and biological subsets of chemical space: Physics-based potentials parametrized from machine learning

Cited by in corpus (66)

• Towards Exact Molecular Dynamics Simulations with Machine-Learned Force Fields
• On-the-fly machine learning force field generation: Application to melting points
• Physics-inspired structural representations for molecules and materials
• GPUMD: A package for constructing accurate machine-learned potentials and performing highly efficient atomistic simulations
• 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
• sGDML: Constructing Accurate and Data Efficient Molecular Force Fields Using Machine Learning
• Machine learning force fields and coarse-grained variables in molecular dynamics: application to materials and biological systems
• Incorporating long-range physics in atomic-scale machine learning
• Accurate molecular polarizabilities with coupled-cluster theory and machine learning
• On the Completeness of Atomic Structure Representations
• Atom-Density Representations for Machine Learning
• Unsupervised machine learning in atomistic simulations, between predictions and understanding
• Fast and Accurate Uncertainty Estimation in Chemical Machine Learning
• Machine-learning interatomic potential for radiation damage and defects in tungsten
• Representations of molecules and materials for interpolation of quantum-mechanical simulations via machine learning
• Active learning of reactive Bayesian force fields: Application to heterogeneous hydrogen-platinum catalysis dynamics
• Feature Optimization for Atomistic Machine Learning Yields A Data-Driven Construction of the Periodic Table of the Elements
• QM7-X: A comprehensive dataset of quantum-mechanical properties spanning the chemical space of small organic molecules
• How to validate machine-learned interatomic potentials
• 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
• A local Bayesian optimizer for atomic structures
• Using Gaussian Process Regression to Simulate the Vibrational Raman Spectra of Molecular Crystals
• Gaussian approximation potentials for body-centered-cubic transition metals
• Error-Controlled Exploration of Chemical Reaction Networks with Gaussian Processes
• PiNN: A Python Library for Building Atomic Neural Networks of Molecules and Materials
• Recursive evaluation and iterative contraction of $N$-body equivariant features
• Data-driven simulation and characterisation of gold nanoparticle melting
• Simple machine-learned interatomic potentials for complex alloys
• Hierarchical Visualization of Materials Space with Graph Convolutional Neural Networks
• Phase Transitions of Zirconia: Machine-Learned Force Fields Beyond Density Functional Theory
• Universal QM/MM Approaches for General Nanoscale Applications
• Equivariant representations for molecular Hamiltonians and N-center atomic-scale properties
• Ranking the information content of distance measures
• Bayesian Force Fields from Active Learning for Simulation of Inter-Dimensional Transformation of Stanene
• 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
• Building machine learning force fields for nanoclusters
• Efficient implementation of atom-density representations
• On Machine Learning Force Fields for Metallic Nanoparticles
• $\alpha$-$\beta$ phase transition of zirconium predicted by on-the-fly machine-learned force field
• Simulating solvation and acidity in complex mixtures with first-principles accuracy: the case of CH${}_3$SO${}_3$H and H${}_2$O${}_2$ in phenol
• A New Kind of Atlas of Zeolite Building Blocks
• CIDER: An Expressive, Nonlocal Feature Set for Machine Learning Density Functionals with Exact Constraints
• Model-free quantification of completeness, uncertainties, and outliers in atomistic machine learning using information theory
• Completeness of Atomic Structure Representations
• Multiscale machine-learning interatomic potentials for ferromagnetic and liquid iron
• Navigating chemical reaction space with a steering wheel
• Compact atomic descriptors enable accurate predictions via linear models
• Indirect Learning of Interatomic Potentials for Accelerated Materials Simulations
• Efficacy of the Radial Pair Potential Approximation for Molecular Dynamics Simulations of Dense Plasmas
• Nonlocal Machine-Learned Exchange Functional for Molecules and Solids
• Machine-learning correction to density-functional crystal structure optimization
• Machine-learning of atomic-scale properties based on physical principles
• Mean-Field Density Matrix Decompositions
• Super-resolution in Molecular Dynamics Trajectory Reconstruction with Bi-Directional Neural Networks
• Segregation, ordering, and precipitation in WTaV-based concentrated refractory alloys
• Construction of Machine Learned Force Fields with Quantum Chemical Accuracy: Applications and Chemical Insights
• Building nonparametric $n$-body force fields using Gaussian process regression
• Accurate Molecular Dynamics Enabled by Efficient Physically-Constrained Machine Learning Approaches
• Machine learning at the atomic-scale
• Modeling refractory high-entropy alloys with efficient machine-learned interatomic potentials: defects and segregation
• Towards Improved Quantum Machine Learning for Molecular Force Fields
• Unified theory of atom-centered representations and message-passing machine-learning schemes