AI RESEARCH

Parameter-free Dynamic Regret: Time-varying Movement Costs, Delayed Feedback, and Memory

arXiv CS.LG

ArXi:2602.06902v2 Announce Type: replace In this paper, we study dynamic regret in unconstrained online convex optimization (OCO) with movement costs. Specifically, we generalize the standard setting by allowing the movement cost coefficients $\lambda_t$ to vary arbitrarily over time.