Regensburg 2025 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

SYMD: Symposium AI-driven Materials Design: Recent Developments, Challenges and Perspectives

SYMD 1: AI-driven Materials Design: Recent Developments, Challenges and Perspectives

SYMD 1.5: Invited Talk

Monday, March 17, 2025, 17:15–17:45, H1

Data-Driven Materials Science — •Miguel Marques — Ruhr University Bochum, Germany

We summarize our recent attempts to discover, characterize, and understand inorganic compounds using novel machine learning approaches. We start by motivating why the search for new materials is nowadays one of the most pressing technological problems. Then we summarize our recent work in using crystal-graph attention neural networks for the prediction of materials properties. To train these networks, we developed a dataset of over 5 million density-functional calculations with consistent calculation parameters. Combining the data and the newly developed networks we have already scanned thousands of structural prototypes spanning a space of several billion materials and identified tens of thousands of theoretically stable compounds. We then discuss how these techniques can be used to discover new materials with tailored properties, using as an example the transition temperature of conventional superconductors. Finally, we speculate which role data-driven research will have in the future of materials science.

Keywords: Machine-learning; Density-functional theory; Conventional superconductors; Big data