Opportunities and Challenges for Machine Learning in Materials Science

Abstract

<jats:p> Advances in machine learning have impacted myriad areas of materials science, such as the discovery of novel materials and the improvement of molecular simulations, with likely many more important developments to come. Given the rapid changes in this field, it is challenging to understand both the breadth of opportunities and the best practices for their use. In this review, we address aspects of both problems by providing an overview of the areas in which machine learning has recently had significant impact in materials science, and then we provide a more detailed discussion on determining the accuracy and domain of applicability of some common types of machine learning models. Finally, we discuss some opportunities and challenges for the materials community to fully utilize the capabilities of machine learning. </jats:p>

Machine learning the quantum-chemical properties of metal–organic frameworks for accelerated materials discovery

“…31 In stark contrast to the existing literature on MOFs, significant progress has been made in the development of ML models that can accelerate the quantum-chemical screening process for a wide range of inorganic and molecular compounds. [32][33][34][35][36][37][38][39] One of the fundamental features underlying much of this work has been the use of high-throughput density functional theory 40 (DFT) workflows to construct large-scale electronic structure-property databases, such as those developed for inorganic solids 29,[41][42][43][44][45][46][47] and molecular systems. [48][49][50][51][52] The synergistic combination of highthroughput DFT databases and ML has led to the discovery of a diverse range of materials with sought-after properties, including efficient organic light-emitting diodes, 53 superhard inorganic materials, 54 and thermally conductive polymers, 55 among many others.…”

Machine learning toward advanced energy storage devices and systems

“…Integrating human knowledge into machine learning (Deng et al, 2020) has achieved functions and performance not available before and facilitated the interaction between human beings and machine learning systems, making machine learning decisions understandable to humans. Beyond the field of computer and data sciences such as computer vision, natural language processing, image recognition, and search engine, machine learning is increasingly used in the field of physics (Carleo et al, 2019;Dunjko and Briegel, 2018), chemistry (Goh et al, 2017;Panteleev et al, 2018), biology (Silva et al, 2019;Zitnik et al, 2019), engineering (Flah et al, 2020;Kim et al, 2018;McCoy and Auret, 2019), and materials science (Morgan and Jacobs, 2020). Besides the above-mentioned disciplines, machine learning technologies have great potentials for addressing the development and management of energy storage devices and systems by significantly improving the prediction accuracy and computational efficiency.…”

Autonomous experimentation systems for materials development: A community perspective

“…The use of ML and AI methods for materials applications is now well established and is the topic of recent reviews. 24,40,[66][67][68][69][70] Their use in accelerating tasks in materials Review research can be broadly classified as learning to ''see'' (e.g., spectral interpretation), learning to ''estimate'' (e.g., surrogate models for predicting outcomes), and learning to ''search'' (e.g., optimization). 71 Many ML predictions of new materials and properties have been confirmed experimentally.…”