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 1.2: Invited Talk

Monday, March 17, 2025, 15:30–16:00, H1

GRACE universal interatomic potential for materials discovery and design — •Ralf Drautz — ICAMS, Ruhr-Universität Bochum, Germany

Universal interatomic potentials parameterize the interaction between all chemical elements in the periodic table simultaneously. In my talk I will introduce the Graph Atomic Cluster Expansion (GRACE). GRACE builds on a complete set of graph basis functions and generalizes equivariant message passing neural networks and other machine learning interatomic potentials. Next, I will discuss the parameterization of GRACE across the periodic table and compare the performance of universal GRACE to element-specific potentials.

The ability to simulate thousands or millions of atoms with complex chemistries for extended time scales opens completely new routes for materials discovery and design. I will demonstrate usage scenarios for widely different materials, chemistries and applications.

Finally, I will focus on limitations of current universal interatomic potentials and suggest steps to overcome these.

Keywords: Machine learning interatomic potentials; Density functional theory; Graph Atomic Cluster Expansion; Universal force field; Atomistic simulation