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 Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-06-26 Page: https://delphi.science/r/arxiv/2606.28307 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…