One of the most exciting tools that have entered the material science toolbox in recent years is machine learning. This collection of statistical methods has already proved to be capable of considerably speeding up both fundamental and applied research. At present, we are witnessing an explosion of works that develop and apply machine learning to solid-state systems. We provide a comprehensive overview and analysis of the most recent research in this topic. As a starting point, we introduce machine learning principles, algorithms, descriptors, and databases in materials science. We continue with the description of different machine learning approaches for the discovery of stable materials and the prediction of their crystal structure. Then we discuss research in numerous quantitative structure–property relationships and various approaches for the replacement of first-principle methods by machine learning. We review how active learning and surrogate-based optimization can be applied to improve the rational design process and related examples of applications. Two major questions are always the interpretability of and the physical understanding gained from machine learning models. We consider therefore the different facets of interpretability and their importance in materials science. Finally, we propose solutions and future research paths for various challenges in computational materials science.

/pdf/recent-advances-and-applications-of-machine-learning-in-55gs85xsws.pdf

Recent advances and applications of machine learning in solid-state materials science

Learning from data has led to paradigm shifts in a multitude of disciplines, including web, text and image search, speech recognition, as well as bioinformatics. Can machine learning enable similar breakthroughs in understanding quantum many-body systems? Here we develop an efficient deep learning approach that enables spatially and chemically resolved insights into quantum-mechanical observables of molecular systems. We unify concepts from many-body Hamiltonians with purpose-designed deep tensor neural networks, which leads to size-extensive and uniformly accurate (1 kcal mol−1) predictions in compositional and configurational chemical space for molecules of intermediate size. As an example of chemical relevance, the model reveals a classification of aromatic rings with respect to their stability. Further applications of our model for predicting atomic energies and local chemical potentials in molecules, reliable isomer energies, and molecules with peculiar electronic structure demonstrate the potential of machine learning for revealing insights into complex quantum-chemical systems.

/pdf/quantum-chemical-insights-from-deep-tensor-neural-networks-266u885pa4.pdf

Quantum-Chemical Insights from Deep Tensor Neural Networks

Advances in solid state white lighting technologies witness the explosive development of phosphor materials (down-conversion luminescent materials). A large amount of evidence has demonstrated the revolutionary role of the emerging nitride phosphors in producing superior white light-emitting diodes for lighting and display applications. The structural and compositional versatility together with the unique local coordination environments enable nitride materials to have compelling luminescent properties such as abundant emission colors, controllable photoluminescence spectra, high conversion efficiency, and small thermal quenching/degradation. Here, we summarize the state-of-art progress on this novel family of luminescent materials and discuss the topics of materials discovery, crystal chemistry, structure-related luminescence, temperature-dependent luminescence, and spectral tailoring. We also overview different types of nitride phosphors and their applications in solid state lighting, including general ...

Down-Conversion Nitride Materials for Solid State Lighting: Recent Advances and Perspectives.

Graph networks are a new machine learning (ML) paradigm that supports both relational reasoning and combinatorial generalization. Here, we develop universal MatErials Graph Network (MEGNet) models ...

/pdf/graph-networks-as-a-universal-machine-learning-framework-for-48fv1bqx0h.pdf

Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals

Recent advances in experimental and computational methods are increasing the quantity and complexity of generated data. This massive amount of raw data needs to be stored and interpreted in order to advance the materials science field. Identifying correlations and patterns from large amounts of complex data is being performed by machine learning (ML) algorithms for decades. Recently, the materials science community started to invest in these methodologies to extract knowledge and insights from the accumulated data. This review follows a logical sequence starting from density functional theory (DFT) as the representative instance of electronic structure methods, to the subsequent high-throughput (HT) approach, used to generate large amounts of data. Ultimately, data-driven strategies which include data mining, screening, and machine learning techniques, employ the data generated. We show how these approaches to modern computational materials science are being used to uncover complexities and design novel materials with enhanced properties. Finally, we point to the present research problems, challenges, and potential future perspectives of this new exciting field.

https://iopscience.iop.org/article/10.1088/2515-7639/ab084b/pdf

From DFT to machine learning: recent approaches to materials science–a review

DOI: 10.1002/aenm.201903242 materials in silico,[19–22] high computational costs and poor scaling still limit their effectiveness in exploring unconstrained chemical spaces and/or complex real-world materials. For instance, highthroughput DFT screening works typically limit the search space to hundreds or, at best, thousands of materials, while DFT simulations of materials are mostly limited to typically less than 1000 atoms, i.e., bulk crystals and isolated molecules. ML therefore offers a solution to the materials exploration problem, making predictions of new materials or properties from existing data, which in turn can drive the generation of more data that can be used to further refine the ML models. Here, we will provide an in-depth, critical review of MLguided design and discovery of energy materials, a field where a novel material with superior performance (e.g., higher energy density, higher energy conversion efficiency, etc.) can have a transformative impact on the urgent global problem of climate change. This review is structured along the steps in a typical workflow for materials ML model building, as shown in Figure 1. The next four sections will provide a concise overview of ML concepts designed to give the reader an appreciation of state-of-the-art techniques as well as resources for building ML models for materials. Section 6 reviews the actual application of ML techniques to the discovery and design of various classes of energy materials, from energy storage (e.g., batteries, fuel cells, etc.) to energy conversion (e.g., thermoelectrics, catalysis, etc.). The final section outlines our perspectives on various challenges and opportunities in ML for energy materials design.

A Critical Review of Machine Learning of Energy Materials

Predicting the stability of crystals is one of the central problems in materials science. Today, density functional theory (DFT) calculations remain comparatively expensive and scale poorly with system size. Here we show that deep neural networks utilizing just two descriptors—the Pauling electronegativity and ionic radii—can predict the DFT formation energies of C3A2D3O12 garnets and ABO3 perovskites with low mean absolute errors (MAEs) of 7–10 meV atom−1 and 20–34 meV atom−1, respectively, well within the limits of DFT accuracy. Further extension to mixed garnets and perovskites with little loss in accuracy can be achieved using a binary encoding scheme, addressing a critical gap in the extension of machine-learning models from fixed stoichiometry crystals to infinite universe of mixed-species crystals. Finally, we demonstrate the potential of these models to rapidly transverse vast chemical spaces to accurately identify stable compositions, accelerating the discovery of novel materials with potentially superior properties. Crystal stability prediction is of paramount importance for novel material discovery, with theoretical approaches alternative to expensive standard schemes highly desired. Here the authors develop a deep learning approach which, just using two descriptors, provides crystalline formation energies with very high accuracy.

/pdf/deep-neural-networks-for-accurate-predictions-of-crystal-56upczs14a.pdf

Deep neural networks for accurate predictions of crystal stability.

Predicting the stability of crystals is one of the central problems in materials science. Today, density functional theory (DFT) calculations are the computational tool of choice to obtain energies of crystals with quantitative accuracy. Despite algorithmic and computing advances, DFT calculations remain comparatively expensive and scale poorly with system size. Here we show that deep neural networks utilizing just two descriptors - the Pauling electronegativity and ionic radii - can predict the DFT formation energies of C3A2D3O12 garnets with extremely low mean absolute errors of 7-8 meV/atom, an order of magnitude improvement over previous machine learning models and well within the limits of DFT accuracy. Further extension to mixed garnets with little loss in accuracy can be achieved using a binary encoding scheme that introduces minimal increase in descriptor dimensionality. Our results demonstrate that generalizable deep-learning models for quantitative crystal stability prediction can be built on a small set of chemically-intuitive descriptors. Such models provide the means to rapidly transverse vast chemical spaces to accurately identify stable compositions, accelerating the discovery of novel materials with potentially superior properties.

/pdf/deep-neural-networks-for-accurate-predictions-of-garnet-31v3vjfolj.pdf

Deep Neural Networks for Accurate Predictions of Garnet Stability

Predicting the properties of a material from the arrangement of its atoms is a fundamental goal in materials science. While machine learning has emerged in recent years as a new paradigm to provide rapid predictions of materials properties, their practical utility is limited by the scarcity of high-fidelity data. Here, we develop multi-fidelity graph networks as a universal approach to achieve accurate predictions of materials properties with small data sizes. As a proof of concept, we show that the inclusion of low-fidelity Perdew–Burke–Ernzerhof band gaps greatly enhances the resolution of latent structural features in materials graphs, leading to a 22–45% decrease in the mean absolute errors of experimental band gap predictions. We further demonstrate that learned elemental embeddings in materials graph networks provide a natural approach to model disorder in materials, addressing a fundamental gap in the computational prediction of materials properties. Multi-fidelity graph networks learn more effective representations for materials from large data sets of low-fidelity properties, which can then be used to make accurate predictions of high-fidelity properties, such as the band gaps of ordered and disordered crystals and energies of molecules.

Weike Ye

Papers

Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals

A Critical Review of Machine Learning of Energy Materials

Deep neural networks for accurate predictions of crystal stability.

Deep Neural Networks for Accurate Predictions of Garnet Stability

Learning properties of ordered and disordered materials from multi-fidelity data