SEP 8, 2026 · PREPRINT
The Exact Time-Uniform Rate Frontier for Stochastic Gradient Descent on Smooth Convex Objectives
arXiv
This is a theoretical analysis establishing convergence rate boundaries for SGD, not a clinical or empirical validation study; it raises mathematical questions about algorithm behavior rather than answering applied or experimental ones.
Reported
Convergence rate ceiling$\sqrt{\log n / n}$
Achievable bound order$h(n)/\sqrt{n}$
Necessary and sufficient condition$\sum_{j=1}^{\infty} \frac{1}{h(2^j)^2} < \…