Signals 4 · free daily AI digest

Silver Rate Is (Almost) Optimal for Gradient Descent Acceleration

Paper recorded by Signals 4 on 2026-09-08 in cs.LG. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-08 on arXiv · recorded by Signals 4 on 2026-09-09

Category: cs.LG · 机器学习 · first seen 2026-09-09

Abstract

We study how far gradient descent (GD) can be accelerated by predetermined nonnegative stepsizes in smooth convex optimization. Writing $p_{\mathrm{sil}}=\log_2(1+\sqrt{2})$, we prove an $Ω\left(n^{-p_{\mathrm{sil}}-O(\sqrt{\log\log n/\log n})}\right)$ non-anytime lower bound. In the anytime setting, every infinite nonnegative schedule has infinitely many horizons with error $Ω\left(n^{-\frac{2p_{

Read on arXiv →

#95 most recent of 215 cs.LG papers we have recorded · ↑ newer: Learning Length-Extrapolatable Recurrent Models · ↓ older: Entropy-Regularized Rank-Masked Policy Optimization for Test-Time Rein
Cite this page: Silver Rate Is (Almost) Optimal for Gradient Descent Acceleration: the #95 most recent of 215 cs.LG papers we have recorded (as of 2026-09-08). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/silver-rate-is-almost-optimal-for-gradient-descent-acceleration.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.LG papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions