Representation of compounds for machine-learning prediction of physical properties

Abstract

The representations of a compound, called "descriptors" or "features", play an essential role in constructing a machine-learning model of its physical properties. In this study, we adopt a procedure for generating a systematic set of descriptors from simple elemental and structural representations. First it is applied to a large dataset composed of the cohesive energy for about 18000 compounds computed by density functional theory (DFT) calculation. As a result, we obtain a kernel ridge prediction model with a prediction error of 0.041 eV/atom, which is close to the "chemical accuracy" of 1 kcal/mol (0.043 eV/atom). The procedure is also applied to two smaller datasets, i.e., a dataset of the lattice thermal conductivity (LTC) for 110 compounds computed by DFT calculation and a dataset of the experimental melting temperature for 248 compounds. We examine the performance of the descriptor sets on the efficiency of Bayesian optimization in addition to the accuracy of the kernel ridge regression models. They exhibit good predictive performances.

11 pages, 5 figures

References in corpus (6)

• A General-Purpose Machine Learning Framework for Predicting Properties of Inorganic Materials
• Distribution of phonon lifetime in Brillouin zone
• Big Data of Materials Science - Critical Role of the Descriptor
• Materials Cartography: Representing and Mining Material Space Using Structural and Electronic Fingerprints
• Compressive sensing as a new paradigm for model building
• Learning physical descriptors for materials science by compressed sensing

Cited by in corpus (41)

• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• DScribe: Library of Descriptors for Machine Learning in Materials Science
• WACSF - Weighted Atom-Centered Symmetry Functions as Descriptors in Machine Learning Potentials
• Developing an improved Crystal Graph Convolutional Neural Network framework for accelerated materials discovery
• Deep Neural Networks for Accurate Predictions of Garnet Stability
• Machine Learning the Voltage of Electrode Materials in Metal-ion Batteries
• Machine learning with force-field inspired descriptors for materials: fast screening and mapping energy landscape
• Materials discovery and properties prediction in thermal transport via materials informatics: a mini-review
• Machine Learning for Predicting Thermal Transport Properties of Solids
• Graph convolutional neural networks as "general-purpose" property predictors: the universality and limits of applicability
• Analytic continuation via domain-knowledge free machine learning
• Materials Representation and Transfer Learning for Multi-Property Prediction
• NOMAD 2018 Kaggle Competition: Solving Materials Science Challenges Through Crowd Sourcing
• Atomic positions independent descriptor for machine learning of material properties
• Transferable and extensible machine learning derived atomic charges for modeling hybrid nanoporous materials
• Matrix- and tensor-based recommender systems for the discovery of currently unknown inorganic compounds
• Group-theoretical high-order rotational invariants for structural representations: Application to linearized machine learning interatomic potential
• Compositional descriptor-based recommender system accelerating the materials discovery
• Toward Accurate Interpretable Predictions of Materials Properties within Transformer Language Models
• IRNet: A General Purpose Deep Residual Regression Framework for Materials Discovery
• Extensible Structure-Informed Prediction of Formation Energy with Improved Accuracy and Usability employing Neural Networks
• Machine Learning - Driven Materials Discovery: Unlocking Next-Generation Functional Materials - A review
• Exploring a potential energy surface by machine learning for characterizing atomic transport
• Genarris: Random Generation of Molecular Crystal Structures and Fast Screening with a Harris Approximation
• Bayesian optimization for computationally extensive probability distributions
• Efficient Estimation of Material Property Curves and Surfaces via Active Learning
• Understanding Magnetic Properties of Actinide-Based Compounds from Machine Learning
• Machine Learning for compositional disorder: A Comparison Between Different Descriptors and Machine Learning Frameworks
• Reconstructing dynamical networks via feature ranking
• Study of Deep Generative Models for Inorganic Chemical Compositions
• Machine learning at the atomic-scale
• Advancing descriptor search in materials science: feature engineering and selection strategies
• A Sampling Strategy in Efficient Potential Energy Surface Mapping for Predicting Atomic Diffusivity in Crystals by Machine Learning
• Decoupling approximation robustly reconstructs directed dynamical networks
• Machine Learning Models for the Lattice Thermal Conductivity Prediction of Inorganic Materials
• Data-Driven Forecasting of Non-Equilibrium Solid-State Dynamics
• Prediction of Carbon Nanostructure Mechanical Properties and Role of Defects Using Machine Learning
• A critical assessment of bonding descriptors for predicting materials properties
• Synergistic Fusion of Multi-Source Knowledge via Evidence Theory for High-Entropy Alloy Discovery
• Materials Screening for the Discovery of New Half-Heuslers: Machine Learning versus Ab Initio Methods
• Combining DFT with ML to study size specific interactions between metal clusters and adsorbates