Exploring the Matterverse with Graph Deep Learning

Prof Ong gave a talk on "Exploring the Matterverse with Graph Deep Learning" at the PSI-K 2022 and ENGE conferences. The matterverse is vast and complex. It comprises the infinite combinations of elements of the periodic table in ordered and disordered arrangements. While ab initio techniques can accurately probe the matterverse, the scope of our exploration has been bottlenecked by their high cost and poor scaling. In this talk, Prof Ong will discuss how machine learning (ML) models trained on large materials datasets have revolutionized our ability to explore the matterverse at unprecedented scales and accuracy. Graph deep learning models today can achieve predictive accuracies that allow us to search vast chemical spaces for novel technological materials. ML interatomic potentials can be used to model complex materials with accuracy at large time and length scales, yielding insights beyond the reach of traditional computational techniques. Finally, we will demonstrate how combining the strengths of graph neural networks with many-body interactions may provide the means to access the matterverse in all its chemical diversity and structural complexity.