Hello! I’m an assistant professor at Zhejiang University working on the foundations of data science and automation.

Bio.   Previously I did my phd at Boston University, advised by Yannis Paschalidis and Ashok Cutkosky. After that I did postdoc at Harvard University hosted by Heng Yang, and at Carnegie Mellon University hosted by Aaditya Ramdas. Long ago I did my undergrad at Tsinghua University.

Research.   My theoretical research focuses on algorithmic problems at the intersection of optimization, statistics, and game theory. My priorities are (i) drawing connections across these areas to derive simple, interpretable, and quantitatively strong algorithms from first principles; and (ii) developing theory that leads, rather than follows, the state-of-the-art practice in data science.

On the practical side, my group is interested in improving the efficiency of AI inference systems via optimiziation modeling. Currently we focus on scheduling problems in agent harnesses, such as context management, task decomposition, and multi-agent orchestration.

Group.   I’m fortunate to lead a warm, supportive and intellectually stimulating research group, which also includes the following members.

Please reach out if you are interested in joining us! We run two separate reading groups that focus on fundamental or timely topics in data science. Topics for Fall’26: high-dimensional probability, and inference-system optimization.

Email address: zhiyuzresearch@gmail.com

CV   Google Scholar   Photo   Talks

Publication

Representative works

Other works

PhD Dissertation.  Temporal Aspects of Adaptive Online Learning: Continuity and Representation