Two New Optimization Papers: Optimal SGD Stepsizes and Stochastic Rayleigh-Ritz Ascent
prof_grimmer · x · 2026-11-02
A math optimization account shared two fresh arXiv papers:
- Exact Theory and Optimal Stepsizes for Stochastic Gradient Descent (arXiv:2610.42048), deriving exact theory and optimal stepsize choices for SGD.
- Stochastic Rayleigh-Ritz Ascent: Global Linear Convergence under a S... (arXiv:2610.42071), proving global linear convergence guarantees.
Both are foundational optimization-theory work; the repost adds no further commentary.
Related event: New paper establishes exact optimal stepsize schedules for SGD(4 posts)→
More from Research
- SemiAnalysis launches GDPx benchmark: best frontier model scores just 12% — rajistics · 2026-11-03
- Anthropic RCT: analysts follow their agent's secretly randomized lean 86% of the time — dl_weekly · 2026-11-03
- Frank Calegari essay: mathematics has changed with AI — mathematicians should act like it — littmath · 2026-11-03
- Rohan Anil's CoreAuto trains MoE across two global clusters over a 25 Gbps link — _arohan_ · 2026-11-03
- LLM4Code 2027 workshop at ICSE Dublin opens CFP with Nov 13 deadline — LingmingZhang · 2026-11-03
- Blogger claims AI refuted the Exponential Time Hypothesis, neared crypto key curves — on paper — LeviTurk · 2026-11-03