A Performance and Cost Assessment of Machine Learning Interatomic Potentials

Abstract

Machine learning of the quantitative relationship between local environment descriptors and the potential energy surface of a system of atoms has emerged as a new frontier in the development of interatomic potentials (IAPs). Here, we present a comprehensive evaluation of ML-IAPs based on four local environment descriptors --- Behler-Parrinello symmetry functions, smooth overlap of atomic positions (SOAP), the Spectral Neighbor Analysis Potential (SNAP) bispectrum components, and moment tensors --- using a diverse data set generated using high-throughput density functional theory (DFT) calculations. The data set comprising bcc (Li, Mo) and fcc (Cu, Ni) metals and diamond group IV semiconductors (Si, Ge) is chosen to span a range of crystal structures and bonding. All descriptors studied show excellent performance in predicting energies and forces far surpassing that of classical IAPs, as well as predicting properties such as elastic constants and phonon dispersion curves. We observe a general trade-off between accuracy and the degrees of freedom of each model, and consequently computational cost. We will discuss these trade-offs in the context of model selection for molecular dynamics and other applications.

References in corpus (37)

• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• On representing chemical environments
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• Deep Potential Molecular Dynamics: a scalable model with the accuracy of quantum mechanics
• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• Moment Tensor Potentials: a class of systematically improvable interatomic potentials
• Machine Learning of Accurate Energy-Conserving Molecular Force Fields
• A Spectral Analysis Method for Automated Generation of Quantum-Accurate Interatomic Potentials
• Assessing the Performance of Recent Density Functionals for Bulk Solids
• DScribe: Library of Descriptors for Machine Learning in Materials Science
• Comparing molecules and solids across structural and alchemical space
• By-passing the Kohn-Sham equations with machine learning
• Active learning of linearly parametrized interatomic potentials
• Machine-learning based interatomic potential for amorphous carbon
• Error estimates for solid-state density-functional theory predictions: an overview by means of the ground-state elemental crystals
• How van der Waals interactions determine the unique properties of water
• Machine learning a general purpose interatomic potential for silicon
• SchNetPack: A Deep Learning Toolbox For Atomistic Systems
• Accelerating crystal structure prediction by machine-learning interatomic potentials with active learning
• Ab initio thermodynamics of liquid and solid water
• Achieving DFT accuracy with a machine-learning interatomic potential: thermomechanics and defects in bcc ferromagnetic iron
• Embedded Atom Neural Network Potentials: Efficient and Accurate Machine Learning with a Physically Inspired Representation
• Data-driven learning of total and local energies in elemental boron
• Extending the Accuracy of the SNAP Interatomic Potential Form
• A Machine Learning Potential for Graphene
• A learning scheme to predict atomic forces and accelerate materials simulations
• Machine-learned multi-system surrogate models for materials prediction
• Non-covalent interactions across organic and biological subsets of chemical space: Physics-based potentials parametrized from machine learning
• Comparison of permutationally invariant polynomials, neural networks, and Gaussian approximation potentials in representing water interactions through many-body expansions
• Accurate Force Field for Molybdenum by Machine Learning Large Materials Data
• Quantum-Accurate Spectral Neighbor Analysis Potential Models for Ni-Mo Binary Alloys and FCC Metals
• Generalized Stacking Fault Energy Surfaces and Dislocation Properties of Silicon: A First-Principles Theoretical Study
• Data-driven Material Models for Atomistic Simulation
• Relating melting trends and elasticity in simple metals: an empirical potential approach
• Automated Calculation of Thermal Rate Coefficients using Ring Polymer Molecular Dynamics and Machine-Learning Interatomic Potentials with Active Learning
• Comparison between classical potentials and ab initio for silicon under large shear
• An Electrostatic Spectral Neighbor Analysis Potential (eSNAP) for Lithium Nitride

Cited by in corpus (168)

• A Universal Graph Deep Learning Interatomic Potential for the Periodic Table
• Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems
• Exceptional piezoelectricity, high thermal conductivity and stiffness and promising photocatalysis in two-dimensional MoSi2N4 family confirmed by first-principles
• Physics-inspired structural representations for molecules and materials
• Machine-learning interatomic potentials for materials science
• Machine learning force fields and coarse-grained variables in molecular dynamics: application to materials and biological systems
• Accelerating first-principles estimation of thermal conductivity by machine-learning interatomic potentials: A MTP/ShengBTE solution
• Towards high-throughput superconductor discovery via machine learning
• Learning Properties of Ordered and Disordered Materials from Multi-fidelity Data
• Deep Potentials for Materials Science
• A practical guide to machine learning interatomic potentials -- Status and future
• Machine Learning Inter-Atomic Potentials Generation Driven by Active Learning: A Case Study for Amorphous and Liquid Hafnium dioxide
• Exploring Phononic Properties of Two-Dimensional Materials using Machine Learning Interatomic Potentials
• Representations of molecules and materials for interpolation of quantum-mechanical simulations via machine learning
• How to validate machine-learned interatomic potentials
• Machine learning in nuclear materials research
• Performance Assessment of Universal Machine Learning Interatomic Potentials: Challenges and Directions for Materials' Surfaces
• A physics-informed operator regression framework for extracting data-driven continuum models
• Machine Learning for Predicting Thermal Transport Properties of Solids
• Atomic-scale origin of the low grain-boundary resistance in perovskite solid electrolytes
• Gaussian approximation potentials for body-centered-cubic transition metals
• Efficient and Accurate Simulations of Vibrational and Electronic Spectra with Symmetry-Preserving Neural Network Models for Tensorial Properties
• Multi-scale Investigation of Chemical Short-Range Order and Dislocation Glide in the MoNbTi and TaNbTi Refractory Multi-Principal Element Alloys
• Ultra-fast interpretable machine-learning potentials
• Recursive evaluation and iterative contraction of $N$-body equivariant features
• Machine learned Force-Fields for an ab-initio Quality Description of Metal-Organic Frameworks
• Explicit Multi-element Extension of the Spectral Neighbor Analysis Potential for Chemically Complex Systems
• Accelerating Finite-temperature Kohn-Sham Density Functional Theory with Deep Neural Networks
• Machine learning for materials discovery: two-dimensional topological insulators
• Machine Learning Interatomic Potential for Simulations of Carbon at Extreme Conditions
• The Evolution of Machine Learning Potentials for Molecules, Reactions and Materials
• Accurate large-scale simulations of siliceous zeolites by neural network potentials
• High thermal conductivity in semiconducting Janus and non-Janus diamanes
• Machine learning accelerated discovery of corrosion-resistant high-entropy alloys
• Bayesian Force Fields from Active Learning for Simulation of Inter-Dimensional Transformation of Stanene
• JARVIS-Leaderboard: A Large Scale Benchmark of Materials Design Methods
• Discrepancies and Error Evaluation Metrics for Machine Learning Interatomic Potentials
• A Perspective on Deep Learning for Molecular Modeling and Simulations
• Quantum-accurate magneto-elastic predictions with classical spin-lattice dynamics
• Short-range order and its impacts on the BCC NbMoTaW multi-principal element alloy by the machine-learning potential
• An Experimentally Driven Automated Machine Learned lnter-Atomic Potential for a Refractory Oxide
• MAISE: Construction of neural network interatomic models and evolutionary structure optimization
• Combining Machine Learning and Many-Body Calculations: Coverage-Dependent Adsorption of CO on Rh(111)
• High-accuracy thermodynamic properties to the melting point from ab initio calculations aided by machine-learning potentials
• Efficient implementation of atom-density representations
• Accurate machine learning force fields via experimental and simulation data fusion
• Lifelong Machine Learning Potentials
• Combining phonon accuracy with high transferability in Gaussian approximation potential models
• Machine learning potentials for multicomponent systems: The Ti-Al binary system
• Dielectric response with short-ranged electrostatics
• Development of a general-purpose machine-learning interatomic potential for aluminum by the physically-informed neural network method
• Transferability of neural network potentials for varying stoichiometry: phonons and thermal conductivity of Mn${}_x$Ge${}_y$ compounds
• Accelerating Atomistic Simulations with Piecewise Machine Learned Ab Initio Potentials at Classical Force Field-like Cost
• Elinvar effect in $\beta -$Ti simulated by on-the-fly trained moment tensor potential
• Efficiency, Accuracy, and Transferability of Machine Learning Potentials: Application to Dislocations and Cracks in Iron
• Thermophysical properties of FLiBe using moment tensor potentials
• Layer-by-layer phase transformation in Ti${}_3$O${}_5$ revealed by machine learning molecular dynamics simulations
• Exploring the robust extrapolation of high-dimensional machine learning potentials
• Automated free energy calculation from atomistic simulations
• Sensitivity and Dimensionality of Atomic Environment Representations used for Machine Learning Interatomic Potentials
• Progress, challenges and perspectives of computational studies on glassy superionic conductors for solid-state batteries
• Quantum lattice dynamics and their importance in ternary superhydride clathrates
• Development of a physically-informed neural network interatomic potential for tantalum
• AENET-LAMMPS and AENET-TINKER: Interfaces for Accurate and Efficient Molecular Dynamics Simulations with Machine Learning Potentials
• Machine-learning potentials for nanoscale simulations of deformation and fracture: example of TiB${}_2$ ceramic
• Enabling robust offline active learning for machine learning potentials using simple physics-based priors
• Combining quantum mechanics and machine-learning calculations for anharmonic corrections to vibrational frequencies
• Multiscale machine-learning interatomic potentials for ferromagnetic and liquid iron
• Lightweight and Effective Tensor Sensitivity for Atomistic Neural Networks
• Compact atomic descriptors enable accurate predictions via linear models
• A Physically-informed Graph-based Order Parameter for the Universal Characterization of Atomic Structures
• Machine-learning potentials enable predictive $\text{and}$ tractable high-throughput screening of random alloys
• UV-Visible Absorption Spectra of Solvated Molecules by Quantum Chemical Machine Learning
• A deep machine learning potential for atomistic simulation of Fe-Si-O systems under Earth's outer core conditions
• Accurate Deep Learning-aided Density-free Strategy for Many-Body Dispersion-corrected Density Functional Theory
• Enhancing the Accuracy of Density Functional Tight Binding Models Through ChIMES Many-body Interaction Potentials
• Innate Dynamics and Identity Crisis of a Metal Surface Unveiled by Machine Learning of Atomic Environments
• Bayesian, frequentist, and information geometric approaches to parametric uncertainty quantification of classical empirical interatomic potentials
• Indirect Learning of Interatomic Potentials for Accelerated Materials Simulations
• Machine Learned Interatomic Potential for Dispersion Strengthened Plasma Facing Components
• Linear Jacobi-Legendre expansion of the charge density for machine learning-accelerated electronic structure calculations
• Through the eyes of a descriptor: Constructing complete, invertible descriptions of atomic environments
• Accelerating the prediction of inorganic surfaces with machine learning interatomic potentials
• Automated determination of grain boundary energy and potential-dependence using the OpenKIM framework
• On the Active Components in Crystalline Li-Nb-O and Li-Ta-O Coatings from First Principles
• Prediction of stable Li-Sn compounds: boosting ab initio searches with neural network potentials
• Measuring transferability issues in machine-learning force fields: The example of Gold-Iron interactions with linearized potentials
• Optimal radial basis for density-based atomic representations
• Lattice dynamics and elastic properties of alpha-U at high-temperature and high-pressure by machine learning potential simulations
• Data efficiency and extrapolation trends in neural network interatomic potentials
• Chemical order transitions within extended interfacial segregation zones in NbMoTaW
• Modeling the Ga/As binary system across temperaturesand compositions from first principles
• A "Magnetic" Machine Learning Interatomic Potential for Nickel
• Constructing and evaluating machine-learned interatomic potentials for Li-based disordered rocksalts
• Machine Learned Interatomic Potentials for Ternary Carbides trained on the AFLOW Database
• Reducing Training Data Needs with Minimal Multilevel Machine Learning (M3L)
• Dynamical properties across different coarse-grained models for ionic liquids
• Modeling Extensive Defects in Metals through Classical Potential-Guided Sampling and Automated Configuration Reconstruction
• Decoding the hidden dynamics of super-Arrhenius hydrogen diffusion in multi-principal element alloys via machine learning
• Neural Networks Potential from the Bispectrum Component: A Case Study on Crystalline Silicon
• Coarse-grained conformational surface hopping: Methodology and transferability
• Understanding the flat thermal conductivity of La2Zr2O7 at ultrahigh temperatures
• Artificial Intelligence for Science in Quantum, Atomistic, and Continuum Systems
• Lattice thermal conductivity and elastic modulus of XN4 (X=Be, Mg and Pt) 2D materials using machine learning interatomic potentials
• Machine-Learned Atomic Cluster Expansion Potentials for Fast and Quantum-Accurate Thermal Simulations of Wurtzite AlN
• Dynamical theory of angle-resolved electron energy loss and gain spectroscopies of phonons and magnons in transmission electron microscopy including multiple scattering effects
• Automated optimization of convergence parameters in plane wave density functional theory calculations via a tensor decomposition-based uncertainty quantification
• Ab initio Canonical Sampling based on Variational Inference
• Proper Orthogonal Descriptors for Efficient and Accurate Interatomic Potentials
• Complex Strengthening Mechanisms in the NbMoTaW Multi-Principal Element Alloy
• Revisit Many-body Interaction Heat Current and Thermal Conductivity Calculation in Moment Tensor Potential/LAMMPS Interface
• Facilitating {\it ab initio} configurational sampling of multicomponent solids using an on-lattice neural network model and active learning
• Multiscale Modeling of Metal/Oxide/Metal Conductive Bridging Random Access Memory Cells: from Ab Initio to Finite Element Calculations
• Probing the state of hydrogen in $\delta$-AlOOH at mantle conditions with machine learning potential
• The JARVIS Infrastructure is All You Need for Materials Design
• Shadow molecular dynamics and atomic cluster expansions for flexible charge models
• Spectral Neural Network Potentials for Binary Alloys
• Exploration of amorphous V${}_2$O${}_5$ as cathode for magnesium batteries
• Does Hessian Data Improve the Performance of Machine Learning Potentials?
• Quasicrystal bulk and surface energies from density functional theory
• Tell machine learning potentials what they are needed for: Simulation-oriented training exemplified for glycine
• Developing Potential Energy Surfaces for Graphene-based 2D-3D Interfaces from Modified High Dimensional Neural Networks for Applications in Energy Storage
• Automated atomistic simulations of dissociated dislocations with ab initio accuracy
• AI-accelerated Materials Informatics Method for the Discovery of Ductile Alloys
• Accelerating the theoretical study of Li-polysulphide adsorption on single-atom catalysts via machine learning approaches
• Efficient moment tensor machine-learning interatomic potential for accurate description of defects in Ni-Al Alloys
• Fast proper orthogonal descriptors for many-body interatomic potentials
• Stability and distortion of fcc-LaH${}_{10}$ with path-integral molecular dynamics
• Critical assessment of machine-learned repulsive potentials for the Density Functional based Tight-Binding method: a case study for pure silicon
• Benchmarking Structural Evolution Methods for Training of Machine Learned Interatomic Potentials
• On the Quality of Uncertainty Estimates from Neural Network Potential Ensembles
• Morphological evolution via surface diffusion learned by convolutional, recurrent neural networks: extrapolation and prediction uncertainty
• Rigorous body-order approximations of an electronic structure potential energy landscape
• EOSnet: Embedded Overlap Structures for Graph Neural Networks in Predicting Material Properties
• Machine learning interatomic potential for predicting the thermal properties of uranium nitride
• Machine learning the electronic structure of matter across temperatures
• Transferability of force fields for 2D silicon (silicene)
• Modeling Chemical Exfoliation of Non-van der Waals Chromium Sulfides by Machine Learning Interatomic Potentials and Monte Carlo Simulations
• Equivariant Tensor Network Potentials
• Predictive power of polynomial machine learning potentials for liquid states in 22 elemental systems
• 2D spectroscopies from condensed phase dynamics: Accessing third-order response properties from equilibrium multi-time correlation functions
• Machine-Learning Potentials Predict Orientation- and Mode-Dependent Fracture in Refractory Diborides
• Machine-Learning Surrogate Model for Accelerating the Search of Stable Ternary Alloys
• Structure and thermodynamics of defects in Na-feldspar from a neural network potential
• Machine learning metallic glass critical cooling rates through elemental and molecular simulation based featurization
• Data Fusion of Deep Learned Molecular Embeddings for Property Prediction
• Transferability of interatomic potentials for germanene (2D germanium)
• A general formalism for machine-learning models based on multipolar-spherical harmonics
• Machine-Learned Potentials for Solvation Modeling
• On Simulating Thin-Film Processes at the Atomic Scale Using Machine Learned Force Fields
• Accelerating Complex Materials Discovery with Universal Machine-Learning Potential-Driven Structure Prediction
• Scalable Training of Neural Network Potentials for Complex Interfaces Through Data Augmentation
• High-pressure melting and elastic behavior of vanadium and niobium based on ab initio and machine learning molecular dynamics
• The effect of normal stress on stacking fault energy in face-centered cubic metals
• Unified theory of atom-centered representations and message-passing machine-learning schemes
• Training Data Set Refinement for the Machine Learning Potential of Li-Si Alloys via Structural Similarity Analysis
• Scalable Data-Driven Basis Selection for Linear Machine Learning Interatomic Potentials
• Compressing local atomic neighbourhood descriptors
• Mode-Dependent Phonon Relaxation in fcc Ni: Insights from Molecular Dynamics Simulations with Frozen-Trajectory Excitations
• Evaluating approaches for on-the-fly machine learning interatomic potential for activated mechanisms sampling with the activation-relaxation technique nouveau
• Interpretable, Physics-Informed Learning Reveals Sulfur Adsorption and Poisoning Mechanisms in 13-Atom Icosahedra Nanoclusters
• Machine-Learned Interatomic Potentials for Predicting Physicochemical Properties of Molten Metal-Salt Systems for Calcium Electrolysis
• Dynamically polarisable force-fields for surface simulations via multi-output classification Neural Networks
• Macro-Dipole-Constrainted Learning of Atomic Charges for Accurate Electrostatic Potentials at Electrochemical Interfaces
• Improved capabilities of the TurboGAP code for radiation induced cascade simulations: an illustration with silicon
• Atomistic mechanisms of phase transitions in all-temperature barocaloric material KPF${}_6$
• Tuning Thermal Conductivity and Electron-Phonon Interactions in Carbon and Boron Nitride Moiré Diamanes via Twist Angle Manipulation
• Validation of moment tensor potentials for fcc and bcc metals using EXAFS spectra