In order to make accurate predictions of material properties, current machine-learning approaches generally require large amounts of data, which are often not available in practice. In this work, an all-round framework is presented which relies on a feedforward neural network, the selection of physically-meaningful features and, when applicable, joint-learning. Next to being faster in terms of training time, this approach is shown to outperform current graph-network models on small datasets. In particular, the vibrational entropy at 305 K of crystals is predicted with a mean absolute test error of 0.009 meV/K/atom (four times lower than previous studies). Furthermore, joint-learning reduces the test error compared to single-target learning and enables the prediction of multiple properties at once, such as temperature functions. Finally, the selection algorithm highlights the most important features and thus helps understanding the underlying physics.
@article{arxiv.2004.14766,
title = {MODNet -- accurate and interpretable property predictions for limited materials datasets by feature selection and joint-learning},
author = {Pierre-Paul De Breuck and Geoffroy Hautier and Gian-Marco Rignanese},
journal= {arXiv preprint arXiv:2004.14766},
year = {2021}
}