Signals 4 · free daily AI digest

When Does Scale-Invariant Optimization Become Unstable? An Exact Schedule Law with Weight Decay

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

Normalization renders large parts of neural networks effectively scale invariant, inducing a hidden feedback loop in which learning-rate schedules and weight decay interact through the parameter norm to control the effective step taken by the optimizer. We show that this interaction is governed by an exact discrete-time law: a single scalar quantity captures all schedule and decay forcing, while n

Read on arXiv →

#98 most recent of 215 cs.LG papers we have recorded · ↑ newer: Nearly Tight Rademacher Bounds for Sparsely Activated Neural Networks · ↓ older: Curriculum Learning as Transport: Understanding Curricula with Wassers
Cite this page: When Does Scale-Invariant Optimization Become Unstable? An Exact Schedule Law with Weight Decay: the #98 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/when-does-scale-invariant-optimization-become-unstable-an-exact-schedule-law-wit.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