Signals 4 · free daily AI digest

Common-Mode Collapse and Recovery in Direct Feedback Alignment

Paper recorded by Signals 4 on 2026-09-25 in cs.LG. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-25 on arXiv · recorded by Signals 4 on 2026-09-28

Category: cs.LG · 机器学习 · first seen 2026-09-28

Abstract

Direct feedback alignment (DFA) trains hidden layers through fixed random projections of output error. With tanh hidden units and independent sigmoid outputs, plain stochastic gradient descent can stall near the loss of a constant predictor of class frequencies. We trace this stall to the error's common mode, the component shared across inputs. An exact mean-covariance decomposition separates a ra

Read on arXiv →

#5 most recent of 310 cs.LG papers we have recorded · ↑ newer: New LoRA Skills Should Read but Never Write · ↓ older: Trust Guided Decision Transformer
Cite this page: Common-Mode Collapse and Recovery in Direct Feedback Alignment: the #5 most recent of 310 cs.LG papers we have recorded (as of 2026-09-25). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/common-mode-collapse-and-recovery-in-direct-feedback-alignment.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