{"id":"arxiv/2602.20567","url":"https://arxiv.org/abs/2602.20567","tags":["cs.LG","math.OC","stat.ML"],"score":0,"title":"Stability and Generalization of Push-Sum Based Decentralized Optimization over Directed Graphs","domain":"cs.LG","source":"arxiv","authors":["Yifei Liang","Yan Sun","Xiaochun Cao","Li Shen"],"reviews":0,"subject":"cs.LG","summary":"Push-Sum-based decentralized learning enables optimization over directed communication networks, where information exchange may be asymmetric. While convergence properties of such methods are well understood, their finite-iteration stability and generalization behavior remain unclear due to structural bias induced by column-stochastic mixing and asymmetric error propagation. In this work, we develop a unified uniform-stability framework for the Stochastic Gradient Push (SGP) algorithm that captures the effect of directed topology. A key technical ingredient is an imbalance-aware consistency…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-02-24","reactions":{}}