My research focuses on online learning and sequential decision-making for building efficient, scalable, and trustworthy machine learning systems. I study both system-driven problems in learning theory and theory-guided algorithms for real-world systems, including LLM serving, Mixture-of-Experts (MoE), and edge AI. A central theme of my work is bridging learning theory with system design to enable intelligent, resource-aware computation.

Openings: I am seeking PhD students starting in Fall 2026 and Research Assistants starting in Spring 2026. If you are interested, please email me your CV and transcripts. For more information, please visit this page.

Bio

I am an Assistant Professor in the Department of Computer Science at City University of Hong Kong. Prior to this, I was a joint postdoc at University of Massachusetts Amherst and California Institute of Technology. I received my Ph.D. in Electrical and Computer Engineering from Carnegie Mellon University, after completing a B.E in Communication Engineering at Nanjing University of Posts and Telecommunications.

News

• Nov 2025: Our work on online multi-LLM selection is accepted to AAAI 2026.
• May 2025: Two papers are accepted to ICML 2025 and one paper is accepted to KDD 2025.
• Jan. 2025: Our work on bandits robust to adversarial attacks is accepted to ICLR 2025.
• Dec. 2024: Our work on robust combinatorial contextual bandits is accepted to INFOCOM 2025.
• Dec. 2024: Our work on heterogeneous multi-agent bandits with hints is accepted to AAAI 2025.