Sepehr Mousavi
PhD Student · Computational Mechanics Group, ETH Zurich
Research Interests
- Operator learning for PDEs. Neural surrogates that learn mappings between function spaces to approximate the solution operators of partial differential equations.
- Learning on arbitrary geometries. Neural operator architectures that are designed to for unstructured point clouds and complex geometries.
- Scientific computing. Numerical methods and large-scale simulations for problems in engineering and the physical sciences.