Machine-learned multi-system surrogate models for materials prediction

Abstract

Surrogate machine-learning models are transforming computational materials science by predicting properties of materials with the accuracy of ab initio methods at a fraction of the computational cost. We demonstrate surrogate models that simultaneously interpolate energies of different materials on a dataset of 10 binary alloys (AgCu, AlFe, AlMg, AlNi, AlTi, CoNi, CuFe, CuNi, FeV, NbNi) with 10 different species and all possible fcc, bcc and hcp structures up to 8 atoms in the unit cell, 15\,950 structures in total. We find that the deviation of prediction errors when increasing the number of simultaneously modeled alloys is less than 1\,meV/atom. Several state-of-the-art materials representations and learning algorithms were found to qualitatively agree on the prediction errors of formation enthalpy with relative errors of $<$2.5\% for all systems.

12 pages, 7 figures

References in corpus (21)

• Deep Learning in Neural Networks: An Overview
• 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
• SchNet - a deep learning architecture for molecules and materials
• 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
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• Constituent Quarks and g_1
• Universal Fragment Descriptors for Predicting Electronic Properties of Inorganic Crystals
• How to represent crystal structures for machine learning: towards fast prediction of electronic properties
• Generating derivative structures: Algorithm and applications
• Efficient Representation of Quantum Many-body States with Deep Neural Networks
• Alchemical and structural distribution based representation for improved QML
• Hierarchical modeling of molecular energies using a deep neural network
• Unified Representation of Molecules and Crystals for Machine Learning
• Atom-Density Representations for Machine Learning
• Deep learning and the Schrödinger equation
• Efficiency of Generalized Regular k-point Grids
• Comment on "Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning"
• Crystal Structure Representations for Machine Learning Models of Formation Energies

Cited by in corpus (30)

• A Performance and Cost Assessment of Machine Learning Interatomic Potentials
• Machine-learning interatomic potentials for materials science
• Rapid Generation of Optimal Generalized Monkhorst-Pack Grids
• Representations of molecules and materials for interpolation of quantum-mechanical simulations via machine learning
• Deep Potential generation scheme and simulation protocol for the Li10GeP2S12-type superionic conductors
• Ab initio vibrational free energies including anharmonicity for multicomponent alloys
• Strategies for the Construction of Machine-Learning Potentials for Accurate and Efficient Atomic-Scale Simulations
• Explicit Multi-element Extension of the Spectral Neighbor Analysis Potential for Chemically Complex Systems
• Efficient Training of ANN Potentials by Including Atomic Forces via Taylor Expansion and Application to Water and a Transition-Metal Oxide
• High-accuracy thermodynamic properties to the melting point from ab initio calculations aided by machine-learning potentials
• Autonomous Investigations over WS${}_2$ and Au{111} with Scanning Probe Microscopy
• Artificial Intelligence in Materials Science and Engineering: Current Landscape, Key Challenges, and Future Trajectorie
• Anharmonic Thermodynamics of Vacancies Using a Neural Network Potential
• In operando active learning of interatomic interaction during large-scale simulations
• Constructing and evaluating machine-learned interatomic potentials for Li-based disordered rocksalts
• Machine Learned Interatomic Potentials for Ternary Carbides trained on the AFLOW Database
• Slippery paraelectric transition metal dichalcogenide bilayers
• Physics-consistent machine learning: output projection onto physical manifolds
• Bypassing the computational bottleneck of quantum-embedding theories for strong electron correlations with machine learning
• Ab initio framework for deciphering trade-off relationships in multi-component alloys
• Machine learning at the atomic-scale
• Free energy of (CoxMn1-x)3O4 mixed phases from machine-learning-enhanced ab initio calculations
• Quantum-informed simulations for mechanics of materials: DFTB+MBD framework
• Equivariant Tensor Network Potentials
• Investigating 3D Atomic Environments for Enhanced QSAR
• Hybrid localized graph kernel for machine learning energy-related properties of molecules and solids
• Entropy-based Active Learning of Graph Neural Network Surrogate Models for Materials Properties
• Trans-dimensional Hamiltonian model selection and parameter estimation from sparse, noisy data
• Gaussian Processes for Finite Size Extrapolation of Many-Body Simulations
• High-throughput spin-bath characterization of spin-defects in semiconductors