{"id":"arxiv/2609.27199","url":"https://arxiv.org/abs/2609.27199","tags":["cs.LG","eess.SY","math.OC"],"score":0,"title":"ZO-COSMO: Index-Free One-Hop Mixing for Decentralized Zeroth-Order Optimization","domain":"cs.LG","source":"arxiv","authors":["Shengjun Zhang","Tingyi Liu","Heng Zhang","Dong Xie"],"reviews":0,"subject":"cs.LG","summary":"Sparse communication in decentralized zeroth-order learning requires compatible peer-state coordinates. We characterize this one-hop condition and develop \\textsf{ZO-COSMO}, coupling two-query estimation with average-preserving masked consensus using $q$ values per active link. Global supports serve all-neighbor mixing; matching updates require agreement only within each pair. We derive a sharp contraction-per-scalar bound within the matching class and convergence guarantees for the core and sparse-momentum updates. At fixed matching, exact moment identities characterize how shared…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-09-23","reactions":{}}