#machine learning
Jan 2026
Small Gradient Norm Regret for Online Convex Optimization
A new problem-dependent regret measure for online convex optimization with smooth losses that strictly refines the existing small loss regret, and that it can be arbitrarily sharper when the losses have vanishing curvature around the hindsight decision.
Wenzhi Gao, Chang He, Madeleine Udell
· arXiv.org · 0 citations