Vince Wang

About Me

I am currently a third-year undergraduate student majoring in Artificial Intelligence at the School of Future Technology, South China University of Technology (SCUT).

My first advisor was Prof. Wei Li. Under his supervision, I developed my core research intuition and methodology. Any of my current and future work is fundamentally built on what I learned from him. He is my academic parent, and I remain deeply grateful to him.

I am currently advised by Prof. Fan Zhang on scientific computing. Specifically, we aim to use deep learning as a relatively low-cost yet high-accuracy alternative to traditional DNS / LES / RANS for high-Reynolds-number fluid simulations. I am now making modest contributions toward this long-term vision.

Research Interests

My research focuses on deep learning for scientific computing, especially neural PDE simulation for physical-field dynamics.

I view neural PDE simulation not merely as architecture design, but as a complete scientific-computing pipeline: how physical states are observed, represented, compressed, and evolved determines what a learned simulator can recover, preserve, and predict.

My long-term goal is to develop state-, observation-, and structure-aware principles for efficient, reliable, and physically faithful scientific simulation under incomplete observations and practical computational budgets.