I am a PhD Candidate at UC San Diego department of Computer Science and Engineering advised by Prof. Rose Yu and a research intern at Google Cloud AI. I received my Bachelor degree in Mathematics from Huazhong University of Science and Technology and Master degree in Data Science from Northeastern University. My research interests are in spatiotemporal learning and physics-guided deep learning. For more details, see my resume and google scholar.
- Investigated deep leaning approaches for forecasting highly nonstationary time series.
- Proposed KNF that achieves SOTA performance on many financial times series datasets.
- Developed neural network surrogate models to inform carbon capture design.
- Implemented mesh-based graph neural networks for CFD simulation surrogate development.
- Developed an ODEs-based method for COVID-19 trajectories forecasting.
- Researched the generalizability of deep learning models on learning non-linear dynamical systems.
- Researched the challenging task of spatiotemporal modeling of nonlinear turbulent flows.
- Developed TF-net that unifies RANS-LES coupling with custom-designed U-net.
- Studied long-term forecast of patients’ Aortic Pressure with deep sequence learning.
- Applied deep sequence models innovatively to the estimation of left ventricular volume.
- Created Healthy Heart Index based on SVM for predicting patients’ survival probabilities.
- Proposed a statistical method based on Beta-Binomial model for systematic teacher evaluation.