Preprint
Aug 2026
On the Iterate Convergence of AdaGrad for Generalized Smooth Convex Optimization
This work constructs a counterexample empirically showing that smoothness alone is not sufficient for the sequential convergence of AdaGrad-type algorithms, and suggesting that additional geometric hypotheses are indispensable for sequential convergence results.
Mathieu Besançon, Tung Le
· 0 citations