How to represent crystal structures for machine learning: towards fast prediction of electronic properties
arXiv:1307.1266 · doi:10.1103/PhysRevB.89.205118
Abstract
High-throughput density-functional calculations of solids are extremely time consuming. As an alternative, we here propose a machine learning approach for the fast prediction of solid-state properties. To achieve this, LSDA calculations are used as training set. We focus on predicting metallic vs. insulating behavior, and on predicting the value of the density of electronic states at the Fermi energy. We find that conventional representations of the input data, such as the Coulomb matrix, are not suitable for the training of learning machines in the case of periodic solids. We propose a novel crystal structure representation for which learning and competitive prediction accuracies become possible within an unrestricted class of spd systems. Due to magnetic phenomena learning on d systems is found more difficult than in pure sp systems.
References in corpus (3)
Cited by in corpus (101)
- Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
- SchNet - a deep learning architecture for molecules and materials
- A General-Purpose Machine Learning Framework for Predicting Properties of Inorganic Materials
- Machine Learning Force Fields
- Quantum-Chemical Insights from Deep Tensor Neural Networks
- Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems
- By-passing the Kohn-Sham equations with machine learning
- Big-Data Science in Porous Materials: Materials Genomics and Machine Learning
- Opportunities and Challenges for Machine Learning in Materials Science
- SchNet: A continuous-filter convolutional neural network for modeling quantum interactions
- SchNetPack: A Deep Learning Toolbox For Atomistic Systems
- Machine Learning Energies of 2 M Elpasolite (ABCD) Crystals
- Machine learning for electronically excited states of molecules
- Efficient and Accurate Machine-Learning Interpolation of Atomic Energies in Compositions with Many Species
- Prediction model of band-gap for AX binary compounds by combination of density functional theory calculations and machine learning techniques
- Alchemical and structural distribution based representation for improved QML
- Resolving transition metal chemical space: feature selection for machine learning and structure-property relationships
- Representation of compounds for machine-learning prediction of physical properties
- Materials Cartography: Representing and Mining Material Space Using Structural and Electronic Fingerprints
- WACSF - Weighted Atom-Centered Symmetry Functions as Descriptors in Machine Learning Potentials
- Unified Representation of Molecules and Crystals for Machine Learning
- Electronic Spectra from TDDFT and Machine Learning in Chemical Space
- Understanding molecular representations in machine learning: The role of uniqueness and target similarity
- Predicting the thermodynamic stability of perovskite oxides using machine learning models
- Perspective on integrating machine learning into computational chemistry and materials science
- Predicting Electronic Structure Properties of Transition Metal Complexes with Neural Networks
- Machine learning with force-field inspired descriptors for materials: fast screening and mapping energy landscape
- Unsupervised machine learning in atomistic simulations, between predictions and understanding
- Atom-Density Representations for Machine Learning
- Machine-learned multi-system surrogate models for materials prediction
- Orbital Graph Convolutional Neural Network for Material Property Prediction
- Constructing first-principles phase diagrams of amorphous LixSi using machine-learning-assisted sampling with an evolutionary algorithm
- Representations of molecules and materials for interpolation of quantum-mechanical simulations via machine learning
- Learning the electronic density of states in condensed matter
- A critical examination of robustness and generalizability of machine learning prediction of materials properties
- Learning physical descriptors for materials science by compressed sensing
- Machine Learning on Neutron and X-Ray Scattering
- A Deep Dive into Machine Learning Density Functional Theory for Materials Science and Chemistry
- Accelerating CALYPSO Structure Prediction by Data-driven Learning of Potential Energy Surface
- The Many-Body Expansion Combined with Neural Networks
- Advances of Machine Learning in Materials Science: Ideas and Techniques
- Predicting the Curie temperature of ferromagnets using machine learning
- Machine Learning for Condensed Matter Physics
- Error-Controlled Exploration of Chemical Reaction Networks with Gaussian Processes
- First-principles interatomic potentials for ten elemental metals via compressed sensing
- Stratified construction of neural network based interatomic models for multicomponent materials
- Metadynamics for Training Neural Network Model Chemistries: a Competitive Assessment
- BIGDML: Towards Exact Machine Learning Force Fields for Materials
- Pattern Learning Electronic Density of States
- Hierarchical Visualization of Materials Space with Graph Convolutional Neural Networks
- Accelerated multiscale mechanics modeling in a deep learning framework
- Accelerating Finite-temperature Kohn-Sham Density Functional Theory with Deep Neural Networks
- Physically inspired deep learning of molecular excitations and photoemission spectra
- Extrapolating quantum observables with machine learning: Inferring multiple phase transitions from properties of a single phase
- Learning molecular energies using localized graph kernels
- Recognizing Local and Global Structural Motifs at the Atomic Scale
- NOMAD 2018 Kaggle Competition: Solving Materials Science Challenges Through Crowd Sourcing
- Atomic positions independent descriptor for machine learning of material properties
- A machine learning-based selective sampling procedure for identifying the low energy region in a potential energy surface: a case study on proton conduction in oxides
- PANNA: Properties from Artificial Neural Network Architectures
- Detection of Topological Materials with Machine Learning
- High-Dimensional Potential Energy Surfaces for Molecular Simulations
- Group-theoretical high-order rotational invariants for structural representations: Application to linearized machine learning interatomic potential
- IRNet: A General Purpose Deep Residual Regression Framework for Materials Discovery
- Machine-learning prediction of infrared spectra of interstellar polycyclic aromatic hydrocarbons
- Extensible Structure-Informed Prediction of Formation Energy with Improved Accuracy and Usability employing Neural Networks
- A New Kind of Atlas of Zeolite Building Blocks
- Permutation-invariant distance between atomic configurations
- Screening and understanding Li adsorption on 2-dimensional metallic materials by learning physics
- Genarris: Random Generation of Molecular Crystal Structures and Fast Screening with a Harris Approximation
- Electronic structure based descriptor for characterizing local atomic environments
- Generating equilibrium molecules with deep neural networks
- Supervised machine learning of ultracold atoms with speckle disorder
- Scalable neural networks for the efficient learning of disordered quantum systems
- Formulation Graphs for Mapping Structure-Composition of Battery Electrolytes to Device Performance
- Efficient first principles based modeling via machine learning: from simple representations to high entropy materials
- A Neural Networks study of the phase transitions of Potts model
- Effects of perturbation order and basis set on alchemical predictions
- Tree based machine learning framework for predicting ground state energies of molecules
- Size and Temperature Transferability of Direct and Local Deep Neural Networks for Atomic Forces
- Efficient Learning of a One-dimensional Density Functional Theory
- Sampling algorithms for validation of supervised learning models for Ising-like systems
- Materials property prediction using symmetry-labeled graphs as atomic-position independent descriptors
- Accelerating the prediction of stacking fault energy by combining ab initio calculations and machine learning
- Machine-learning of atomic-scale properties based on physical principles
- Machine learning phases and criticalities without using real data for training
- Supervised learning of few dirty bosons with variable particle number
- Efficient prediction of 3D electron densities using machine learning
- Supervised learning of random quantum circuits via scalable neural networks
- Training-free hyperparameter optimization of neural networks for electronic structures in matter
- Accelerating structure search using atomistic graph-based classifiers
- Machine Learning S-Wave Scattering Phase Shifts Bypassing the Radial Schrödinger Equation
- Ab initio machine learning in chemical compound space
- Machine learning for accelerated bandgap prediction in strain-engineered quaternary III-V semiconductors
- Predicting Quantum Potentials by Deep Neural Network and Metropolis Sampling
- Computational Design of Stable and Highly Ion-conductive Materials using Multi-objective Bayesian Optimization: Case Studies on Diffusion of Oxygen and Lithium
- Ollivier persistent Ricci curvature (OPRC) based molecular representation for drug design
- Stacked Generalization Approach to Improve Prediction of Molecular Atomization Energies
- Combining DFT with ML to study size specific interactions between metal clusters and adsorbates
- Machine learning approach for vibronically renormalized electronic band structures
- Machine learning, phase stability, and disorder with the Automatic Flow Framework for Materials Discovery