DeMuon: A Decentralized Muon for Matrix Optimization over Graphs 文章

ArXiv CS.AI2026-06-03NEWSen作者: Chuan He, Shuyi Ren, Jingwei Mao, Erik G. Larsson
查看原文 →
关系图谱

DeMuon: A Decentralized Muon for Matrix Optimization over Graphs · 相关技术

相关技术

transformersgradient trackingNewton-Schulz iterationsMuonDeMuon