Hi, welcome to my page! I am a Research Scientist at Meta. Prior to this, I was a postdoctoral associate at the Department of Electrical and Computer Engineering, University of Minnesota, mentored by Prof. Mingyi Hong and Prof. Shuzhong Zhang. I obtained my Ph.D. degree in Applied Mathematics at the Department of Mathematics, UC Davis, advised by Prof. Krishna Balasubramanian and Prof. Shiqian Ma. Prior to this, I received my B.S. degree in Mathematics from Zhejiang University. My CV is here.
My research interests lie at the optimization problems arising in machine learning, operations research and other applications. Specifically, I worked or am working on the following topics:
- Computational complexities for gradient-based, gradient-free (zeroth-order) and primal-dual algorithms for solving nonconvex optimizations and optimization on Riemannian manifolds.
- Convergence theory for deterministic and stochastic minimax/bilevel problems with applications in operations research and machine learning.
- Distributed optimization algorithms, including decentralized and federated learning.
- Theories for reinforcement learning algorithms, especially policy-based methods, and their applications in large language model (LLM) alignments.
- Efficient pre-training and fine-tuning of large language models (LLMs).
News
August 2025: Starting my new job as Research Scientist at Meta!
February 2025: paper “Problem-Parameter-Free Decentralized Nonconvex Stochastic Optimization” is accepted by Pacific Journal of Optimization.
- January 2025: One paper is accepted by ICLR 2025 (Spotlight). Congratulations to all my collaborators!
- Joint Demonstration and Preference Learning Improves Policy Alignment with Human Feedback
January 2025: Paper “Riemannian Bilevel Optimization” is accepted by Journal of Machine Learning Research!
November 2024: Paper “A Riemannian ADMM” is accepted by Mathematics of Operations Research!
- September 2024: Two papers are accepted by NeurIPS 2024. Congratulations to all my collaborators!
- Getting More Juice Out of the SFT Data: Reward Learning from Human Demonstration Improves SFT for LLM Alignment
- SLTrain: a sparse plus low-rank approach for parameter and memory efficient pretraining
August 2024: I’m very happy to receive the INFORMS Computing Society Prize!
August 2024: Paper “Zeroth-order Riemannian Averaging Stochastic Approximation Algorithms” is accepted by SIAM Journal on Optimization!
July 2024: A new grant “Bi-Level Optimization for Hierarchical Machine Learning Problems: Models, Algorithms and Applications” is awarded from NSF. I’m excited to be the co-PI of this project with Prof Hong!
- May 2024: Paper “Revisiting Zeroth-Order Optimization for Memory-Efficient LLM Fine-Tuning: A Benchmark” is accepted by ICML 2024. Congratulations to all my collaborators!