New research improves last-iterate convergence bounds for SGD
A new paper studies last-iterate convergence bounds for stochastic gradient descent on convex and beta-smooth loss functions, reporting improved bounds both with and without replacement.
Read the full story at arXiv stat.ML ↗
Timeline · 1 report
- 2026-11-13 05:00 · arXiv stat.ML
Improved last-iterate bounds for SGD with and without replacement