My research focuses on answering the question: “How to bond AI algorithms with physical dynamics to simulate the Way How the real world works.” Specifically, I work on applying machine learning methods in conjunction with numerical schemes for solving partial differential equations (PDEs). Currently, I am a Ph.D. student in Nils Thuerey‘s group, at Technical University of Munich. I received my M.Eng. degree from Zhejiang University, China in June 2023. More info can refer to my Home Page

Publication [Google Scholar]

Publication after joining:

  • Wei H, List B, Thuerey N ReViT: Rotational-equivariant Vision Transformers for Neural PDE Solvers, ICML 2026, Oral ☛ [Project]
  • Wei H, Franz A, List B, Thuerey N INC: An Indirect Neural Corrector for Auto-Regressive Hybrid PDE Solvers, NeurIPS 2025 ☛ [Project]
  • Franz A, Wei H, Guastoni L, et al. PICT–A Differentiable, GPU-Accelerated Multi-Block PISO Solver for Simulation-Coupled Learning Tasks in Fluid Dynamics[J]. JCP, 2025.☛ [Project]

Teaching

    • Summer 2024, 2026: Seminar – Deep Learning in Computer Graphics
    • Summer 2025, 2026: Seminar – Deep Learning in Physics
    • Winter 2024, 2025: Computer Games Laboratory

Bachelor’s / Master’s Theses

If you’re interested in a Bachelor’s or Master’s thesis related to my work, please contact me.

Contact

E-mail: hao.wei (at) tum (dot) de
Phone: +49.89.289.1736
Fax: +49.89.289.17376
Room: 02.13.039