Accelerating Materials Science with Automation, Data and Machine Learning
Such large databases of computed materials properties have proved invaluable in the screening of vast chemical spaces for novel materials discoveries.
In this talk, I will demonstrate how the application of deep learning techniques can enhance the value that can be extracted from large-scale quantum mechanical calculations, from the prediction of entirely novel technological materials, to quantifying fundamental relationships between chemically intuitive descriptors and crystal stability, to accessing time and length scales beyond first principles approaches.
I will also provide some perspectives on future challenges and opportunities in the application of deep learning techniques in materials science.