Distributed Optimization on Matrix‐Weighted Networks0 PRE AI DOI:10.1002/oca.3212OriginalOriginal request for helpShareSaveAbstract En