Improved last-iterate bounds for SGD with and without replacement
stat.ML, cs.LG, math.OC
Submitted: 2026-11-10
Updated: 2026-11-10
Terminology
Sources
- Large-Step Last-Iterate Convergence of Random-Reshuffling Stochastic Gradient Descent
- Bias-Optimal Bounds for Stochastic Gradient Descent
- How AI settled the complexity of the oldest SGD algorithm
- Last-Iterate Complexity of SGD for Convex and Smooth Stochastic Problems
- A Shadow Sequence Technique for Last-Iterate Convergence Analysis of the Random Reshuffling Method: Relaxed Stepsize Conditions and Beyond
- Random Reshuffling for Smooth Convex Optimization: Dominates Stochastic Gradient Descent
- A Resolution of the SS--RS--GD Inequalities
- Near-Optimal Last-Iterate Convergence of SGD in the Smooth Interpolation Regime
- Optimistic Rates for Learning with a Smooth Loss
- When the noncommutative AM-GM inequality holds
Related papers
- Uniform-in-Time Weak Propagation-of-Chaos in Shallow Neural Networks
- Diffusion Models Observe Only Gradients: A Geometric Perspective on Score Matching Errors
- Quantifying Error Propagation and Model Collapse in Diffusion Models
- Behavior of prediction performance metrics with rare events
- Statistical Inference and Learning for Shapley Additive Explanations (SHAP)
- Control Variate Design for Score-Function Gradient Estimators in Importance-Weighted Variational Inference