Zhiyuan Fan

Environment Evolution for Terminal Agents

Zhiyuan Fan, Tinghao Yu, Yuanjun Cai, et al.

Toward Scalable Terminal Task Synthesis via Skill Graphs

Zhiyuan Fan, Tinghao Yu, Yuanjun Cai, et al.

Online Learning and Equilibrium Computation with Ranking Feedback

Mingyang Liu, Yongshan Chen, Zhiyuan Fan, et al.

This paper studies online learning when the learner only receives ranking feedback (like "action A is better than B") instead of numeric scores, which is more practical for human feedback and privacy-sensitive applications. The authors show that learning with instantaneous rankings is fundamentally impossible, but develop algorithms that achieve good performance when utilities change slowly or when using time-averaged rankings. Their approach enables multiple players in games to reach approximate equilibrium through repeated play.