About me

I am an Assistant Professor in the Department of Applied Mathematics at The Hong Kong Polytechnic University in Hong Kong.

My research lies at the intersection of reinforcement learning, diffusion models for generative AI, stochastic control, and financial engineering. I develop continuous-time reinforcement learning and stochastic control methods for learning and decision-making under uncertainty, with an emphasis on theoretical guarantees, model-free algorithms, and empirical validation.

I received my Ph.D. in Industrial Engineering and Operations Research from Columbia University in 2024, where I was advised by Professor Xun Yu Zhou. I also hold an M.S. in Operations Research from Columbia University and a B.S. in Mathematics and Applied Mathematics from Zhejiang University.

News

• November 2026. Upcoming talk at the INFORMS Annual Meeting (San Francisco).
• July 2026. Upcoming talk at the Mathematical Control Theory Conference: The 7th Symposium in Memory of Professor Xunjing Li (Huzhou).
• July 2026. Upcoming talk at the International Conference on Optimal Control and Machine Learning (Guiyang).
• June 2026. Upcoming talk at the Numerical Analysis in Artificial Intelligence (Guangzhou).
• June 2026. Upcoming talk at the Quantitative Finance Conference (NUS Singapore).

Research Overview

My current research focuses on three closely connected directions:

• Diffusion models and generative AI. I study how reinforcement learning and stochastic control can improve diffusion-based generative modeling. My recent work develops adaptive and reusable control methods for diffusion sampling and inverse problems, aiming to improve sampling efficiency while remaining compatible with pretrained diffusion models.

• Financial engineering and decision-making under uncertainty. I develop reinforcement learning and stochastic control methods for financial decision problems, including portfolio optimization, asset-liability management, and dynamic investment. This line of work connects continuous-time theory with empirical validation using real financial data.

• Continuous-time reinforcement learning and stochastic control. I study model-free reinforcement learning algorithms for continuous-time stochastic control problems, including linear-quadratic control and actor-critic methods with exploration. This line of work develops algorithms with theoretical performance guarantees and provides methodological foundations for applications in finance, generative modeling, and other stochastic systems.

My work combines theoretical analysis with empirical validation, with the goal of developing reliable learning-based methods for complex stochastic systems.

Prospective Students

I am looking for self-motivated Ph.D. students and research assistants interested in reinforcement learning, diffusion models for generative AI, stochastic control, and financial engineering. Interested applicants are welcome to email me with a CV, transcript, and a brief description of research interests.

Research Support

My research is supported by the PolyU Start-up Fund for New Recruits.

For more details, please see my Publications and CV.