AI RESEARCH
Optimal Projection-Free Adaptive SGD for Matrix Optimization
arXiv CS.LG
•
ArXi:2604.02505v1 Announce Type: cross Recently, Jiang developed Leon, a practical variant of One-sided Shampoo [Xie, 2025a, An, 2025] algorithm for online convex optimization, which does not require computing a costly quadratic projection at each iteration. Unfortunately, according to the existing analysis, Leon requires tuning an additional hyperparameter in its preconditioner and cannot achieve dimension-independent convergence guarantees for convex optimization problems beyond the bounded gradients assumption.