I am currently a quantitative researcher at Citadel LLC.
Before joining Citadel, I was a Ph.D. student at Stanford University. My Ph.D. research centered on physics-based machine learning for inverse problems in scientific computing. I developed the open-source software ADCME.jl in Julia and C++ for high-performance inverse modeling using automatic differentiation. Specifically, I have developed novel physics-based machine learning algorithms and software packages based on ADCME.jl for solving inverse problems in stochastic processes, solid mechanics, geophysics and fluid dynamics. One highlight of my research is combining neural networks with numerical solvers for PDEs, which enables data-driven modeling with physics knowledge.
- Data-driven inverse modeling;
- Automatic differentiation;
- Numerical partial differential equations.
To have an overview of my research, read my slides on Machine Learning for Inverse Problems in Computational Engineering.
I defended my thesis on 4/22/2021; read my oral defense slide on Machine Learning for Computational Engineering.