Research
I am broadly interested in the science of AI: understanding the principles that explain the behavior of modern learning systems. A central question motivating my research is: Why do simple algorithms and ideas often work surprisingly well in practice, and what fundamental principles are behind it?
My current research focuses on the mathematical foundations of modern machine learning, especially large-scale overparameterized models. I study how loss landscapes, training dynamics, and optimizer design shape their behavior.
Recent and Selected Papers
View All Papers* denotes equal contribution or alphabetical ordering
Service
Organizer
- IFDS Seminar Co-organizer: Fall 2025 (with Libin Zhu), Winter 2026, and Spring 2026 (with Weihang Xu).
Teaching
- CPS590.04 Machine Learning Algorithms, Spring 2021 @ Duke, TA.
- CPS330 Design and Analysis of Algorithms, Fall 2020 @ Duke, TA.
- CPS330 Design and Analysis of Algorithms, Spring 2020 @ Duke, TA.
Conference
- Area Chair: ICLR (2027).
- Reviewer: ICML, ICLR, NeurIPS, COLT, STOC, AAAI, AISTATS, CVPR, ALT.
Journal
- Reviewer: JMLR, Mathematical Programming, TPAMI, JASA, TMLR.