phenomenon · first seen 5 Oct 2026
last-iterate convergence
Convergence guarantees for the final iterate of a learning dynamic rather than an averaged iterate.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Fast Last-Iterate Convergence in Zero-Sum Markov Games with Bandit FeedbackPapers · 5 Oct 2026 · Yuheng Zhang
- Linear Last-Iterate Convergence Rates for Optimistic Learning in Harmonic GamesPapers · 13 Oct 2026 · John Lazarsfeld, Anas Barakat, Jacob Abernethy, Vidya Muthukumar, Georgios Pilio…
- Optimal Last-Iterate Bounds for Nonexpansive Stochastic ApproximationPapers · 10 Nov 2026 · Ramanan Tamizholi, Siddharth Chandak, Rahul Singh
Latest sources
- Optimal Last-Iterate Bounds for Nonexpansive Stochastic ApproximationPapers · 10 Nov 2026 · Ramanan Tamizholi, Siddharth Chandak, Rahul Singh
- Linear Last-Iterate Convergence Rates for Optimistic Learning in Harmonic GamesPapers · 13 Oct 2026 · John Lazarsfeld, Anas Barakat, Jacob Abernethy, Vidya Muthukumar, Georgios Pilio…
- Fast Last-Iterate Convergence in Zero-Sum Markov Games with Bandit FeedbackPapers · 5 Oct 2026 · Yuheng Zhang