No article text was recorded for this version.
last iterate suboptimality
Version 1 of 120 papers25 conceptsfrozen 2026-10-04
Papers in this version
- Blog: Tight analyses for subgradient descent I: Lower bounds
- Blog: Tight analyses for non-smooth stochastic gradient descent
- Blog: The Role of Momentum Parameters in the Optimal Convergence of Adaptive Polyak's Heavy-ball Methods
- Blog: Making the Last Iterate of SGD Information Theoretically Optimal
- Blog: Making the Last Iterate of SGD Information Theoretically Optimal
- Blog: Optimal Stochastic and Online Learning with Individual Iterates
- Blog: Revisiting the Last-Iterate Convergence of Stochastic Gradient Methods
- Blog: Gradient Descent Averaging and Primal-dual Averaging for Strongly Convex Optimization
- Blog: Improved Convergence in High Probability of Clipped Gradient Methods with Heavy Tailed Noise
- Blog: Gradient Descent’s Last Iterate is Often (slightly) Suboptimal
- Blog: A Lyapunov analysis for accelerated gradient methods: from deterministic to stochastic case
- Blog: On Almost Sure Convergence Rates of Stochastic Gradient Methods
- Blog: Lipschitz-free Projected Subgradient Method with Time-varying Step-size
- Blog: Stochastic Gradient Descent for Non-smooth Optimization: Convergence Results and Optimal Averaging Schemes
- Blog: Stab-SGD: Noise-Adaptivity in Smooth Optimization with Stability Ratios
- Blog: Statistical Sparse Online Regression: A Diffusion Approximation Perspective
- Blog: Convergence Rate of the Last Iterate of Stochastic Proximal Algorithms
- Blog: Open Problem: Tight Convergence of SGD in Constant Dimension
- Blog: Optimal Regularized Dual Averaging Methods for Stochastic Optimization
- Blog: Making Gradient Descent Optimal for Strongly Convex Stochastic Optimization
History & citation
Current · Permanent link
- This direction split from last-iterate convergence gap (50% overlap)
- This direction split from Optimal Convergence Rates for the Last Iterate of SGD (60% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_last_iterate_suboptimality_a7b261e7,
title = {last iterate suboptimality},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/last-iterate-suboptimality/rdr_a7b261e7f6464585a5e14e7959e85218},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-22},
}This exact version (version 1, frozen text):
@misc{lacuna_last_iterate_suboptimality_a7b261e7_v1,
title = {last iterate suboptimality},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/last-iterate-suboptimality/rdr_a7b261e7f6464585a5e14e7959e85218/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-22},
}