Scientific Computing, Data-driven Methods for Partial Differential Equations, Finite Element Methods, Approximation Theory of Transformers, PyTorch contributor.
About
Shuhao Cao is a computational mathematician working on how to solve partial differential equations, which serve as "the most powerful tool humanity has ever created for making sense of the material world". More recently, Shuhao is dedicated to bring together the design of Transformer neural network architectures in GPT and the theory of partial differential equations to help the PDE solver become faster and more accurate. Back in high school, inspired by Kip Thorne's book Black Holes & Time Warps: Einstein's Outrageous Legacy, Shuhao decided that he wanted to be a scientist who solves equations. Aside from research, Shuhao likes playing with mathematical puzzles and solving problems such as those in Putnam and Math Olympiad competitions. In his spare time, Shuhao likes jogging, and occasionally runs long distances. Shuhao also loves to play a board game called Go (Weiqi in Chinese, Igo in Japanese, Baduk in Korean). Students interested in scientific computing, solving differential equations, and writing machine learning codes are welcome to come by to discuss potential research opportunities, or simply play a match of Go.
We use cookies to provide and improve our services. By using our site, you consent to cookies.
