{"id":"arxiv/2606.28307","url":"https://arxiv.org/abs/2606.28307","tags":["math.OC","cs.LG"],"score":0,"title":"Second-Order KKT Guarantees for Bregman ADMM in Nonconvex and Non-Lipschitz Optimization","domain":"math.OC","source":"arxiv","authors":["Shuang Li","Zhihui Zhu","Qiuwei Li"],"reviews":0,"subject":"math.OC","summary":"We analyze Bregman ADMM for nonconvex linearly constrained problems under two-sided relative smoothness, a condition that replaces the standard Lipschitz gradient assumption with a Hessian comparison relative to a Bregman kernel. This setting covers polynomial objectives arising in matrix and tensor models for which a global Lipschitz-gradient constant need not exist. We show that on an invariant open state-space domain, one iteration of Bregman ADMM defines a smooth primal--dual fixed-point map whose strict-saddle KKT points are unstable fixed points; consequently, from random…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-06-26","reactions":{}}