Screening billions of candidates for solid lithium-ion conductors: A transfer learning approach for small data

Abstract

Machine learning (ML) methods have the potential to revolutionize materials design, due to their ability to screen materials efficiently. Unlike other popular applications such as image recognition or language processing, large volumes of data are not available for materials design applications. Here, we first show that a standard learning approach using generic descriptors does not work for small data, unless it is guided by insights from physical equations. We then propose a novel method for transferring such physical insights onto more generic descriptors, allowing us to screen billions of unknown compositions for Li-ion conductivity, a scale which was previously unfeasible. This is accomplished by using the accurate model trained with physical insights to create a large database, on which we train a new ML model using the generic descriptors. Unlike previous applications of ML, this approach allows us to screen materials which have not necessarily been tested before (i.e., not on ICSD or Materials Project). Our method can be applied to any materials design application where a small amount of data is available, combined with high details of physical understanding.

Predicting the state of charge and health of batteries using data-driven machine learning

“…The combined database and machine learning approach have been applied to design and predict the material properties of electrodes such as voltage, crystallinity and chemical stability, from atomic scale to mesoscale 83,[94][95][96][97][98][99] . In addition, such an approach has been applied to design new liquid electrolytes and additives [100][101][102][103][104][105] , and solid-state electrolytes with fast Li-ion transport [106][107][108] and mechanical 82 properties. Such computational techniques provide an opportunity for exploring material properties at a lower cost and accelerating the material discovery processes.…”

Machine learning: Accelerating materials development for energy storage and conversion

“…Moreover, not limited to perovskites, τ can also estimate the stability of perovskite‐like structures. Ekin et al demonstrated that the standard ML approach cannot establish suitable models from small data. To solve the problem of limited data, they created a transfer learning approach which combines structural models and elemental models.…”

“…High‐throughput screening methods have been used to explore ideal solid electrolytes. For small training data, Ekin et al revealed a kind of transfer learning methods to screen potential solid lithium‐ion conductors. In order to reduce the consumption of DFT calculations, Sokseiha and the coworkers proposed a new predictor related with high ionic conductivity that can be used in high‐throughput screening.…”

iQSPR in XenonPy: A Bayesian Molecular Design Algorithm

“…The sampling efficiency of iQSPR-X is highly influenced by the reliability of the evaluator that predicts the material properties for any given chemical structure. [18,[34][35][36][37][38][39][40][41] In this study, we applied a specific type of transfer learning using pre-trained neural networks. XenonPy currently provides 140,000 pretrained neural networks for the prediction of physical, chemical, electronic, thermodynamic, and mechanical properties of small organic molecules, polymers, and inorganic crystalline materials, with models for 15, 18, and 12 properties of these material types, respectively.…”

“…Other studies have also shown promising applications of transfer learning in materials informatics. [18,[34][35][36][37][38][39][40][41] In this study, we applied a specific type of transfer learning using pre-trained neural networks. For a target property, a neural network pre-trained on proxy properties is available in the library, where the source datasets are sufficiently large.…”