AI Paper Atlas

Research map / Learning theory and algorithms / Kernels and regression theory

Scaling Laws of SignSGD in Linear Regression: When Does It Outperform SGD?

Jihwan Kim, Dogyoon Song, Chulhee Yun

ICLR 2026 · Learning theory and algorithms › Kernels and regression theory · approach: High-dimensional asymptotic generalization · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace SGD with signSGDlinear PLRF model · PLRF (alpha=0.4, beta=0.8)compute-optimal slopeImproved
replace SGD with signSGDlinear PLRF model · PLRF (alpha=0.7, beta=1.1)compute-optimal slopeImproved
replace SGD with AdamWTransformer · OpenWebTextcompute-optimal loss exponent: -0.005 -> -0.021Improved
add warmup-stable-decay schedule instead of constant learning ratesignSGD · PLRF (alpha=1.0, beta=0)compute-optimal slopeImproved
replace SGD with signSGD in regimes with small noise bottlenecklinear PLRF model · PLRF phases Aa, Ab, Ac, Ba, I, IIcompute-optimal slopeNo clear effect
use maximal instead of optimal learning rate (ablation)signSGD · PLRFcompute-optimal slopeHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy