Timon Rabczuk

MR-GVNO achieves millisecond-level full-field inference for complex plate structures without the need for labeled data, revolutionizing rapid response predictions in engineering applications.

HAMNO outperforms traditional neural operators by effectively balancing local and global information, achieving superior predictive accuracy in complex dynamical systems.

Dmsh achieves globally conforming all-quadrilateral meshes without the need for post hoc corrections, revolutionizing the automation of mesh generation in complex geometries.