Universal fragment descriptors for predicting properties of inorganic crystals

By Olexandr Isayev, Corey Oses, Cormac Toher, Eric Gossett, Stefano Curtarolo, Alexander Tropsha. 2016

599 citations recorded by the source corpus.

Although historically materials discovery has been driven by a laborious trial-and-error process, knowledge-driven materials design can now be enabled by the rational combination of Machine Learning methods and materials databases. Here, data from the AFLOW repository for ab initio calculations is combined with Quantitative Materials Structure-Property Relationship models to predict important properties: metal/insulator classification, band gap energy, bulk/shear moduli, Debye temperature and heat capacities. The prediction’s accuracy compares well with the quality of the training data for virtually any stoichiometric inorganic crystalline material, reciprocating the available thermomechanical experimental data. The universality of the approach is attributed to the construction of the descriptors: Property-Labelled Materials Fragments. The representations require only minimal structural input allowing straightforward implementations of simple heuristic design rules. Machine learning methods can be useful for materials discovery; however certain properties remain difficult to predict. Here, the authors present a universal machine learning approach for modelling the properties of inorganic crystals, which is validated for eight electronic and thermomechanical properties.

Subfields: Inorganic Chemistry, Materials Chemistry

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Inorganic Chemistry

In Materials Chemistry

Citations in the corpus

Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals (2018, methodology): “54 The MAEs of the JARVIS-ML models 53 for formation energy, band gap, bulk moduli and shear moduli are 0.12 eV atom − 1 , 0.32 eV, 10.5 GPa and 9.5 GPa, respectively, while the MAEs of AFLOW-ML models 54 for band gap, bulk moduli and shear moduli are 0.35 eV, 8.68 GPa and 10.62 GPa, respectively.”
Atomistic Line Graph Neural Network for improved materials property predictions (2021, background): “…in the development of GNN architectures for predicting material properties such as SchNet 10 , Crystal Graph Convolutional Neural Networks (CGCNN) 15 , MatErials Graph Network (MEGNet) 16 , improved Crystal Graph Convolutional Neural Networks (iCGCNN) 17 , OrbNet 18 and similar variants 19-31 .”
A perspective on conventional high-temperature superconductors at high pressure: Methods and materials (2019, unknown): “Other extraordinary examples in the solid-state community is the one led by Curtarolo and his team that exploits repository for ab initio calculations combined with Quantitative Materials Structure-Property Relationship models to predict properties such as metal/insulator classification, band gap en”