No AI summary available for this article.
Why It Matters
Heavy-tailed noise has been widely observed in modern machine learning, motivating the use of methods like gradient clipping and normalization.
Provenance
Discovered via ArXiv and published by ArXiv.
Key Claims
Original description
Heavy-tailed noise has been widely observed in modern machine learning, motivating the use of methods like gradient clipping and normalization. While these methods are well understood in centralized settings, much less is known in decentralized ones, where applying a nonlinearity to local gradients affects both optimization and consensus. Recent works on decentralized non-convex optimization have studied both clipping and normalization under heavy-tailed noise, with clipping yielding suboptimal rates and normalization needing local momentum or mini-batches to converge. This raises the question...
Discovered via ArXiv
Research papers and preprints from arXiv.
Publisher: arxiv.org
ID: http://arxiv.org/abs/2610.10527v1 · Indexed about 2 hours ago