{"id":"arxiv/2610.16765","url":"https://arxiv.org/abs/2610.16765","tags":["math.OC"],"score":0,"title":"Globally Convergent Block Coordinate Descent Algorithms for Semi-Nonnegative Independent Component Analysis","domain":"math.OC","source":"arxiv","authors":["Zhenyu Cai","Wentao Ding","Jianze Li","Wen Huang"],"reviews":0,"subject":"math.OC","summary":"We propose a subgradient-based block coordinate descent (BCD-SubG) framework for nonconvex optimization over products of probability simplices and a closed feasible set, motivated by semi-nonnegative independent component analysis (Semi-NNICA). Elementary column transformations are used to preserve simplex feasibility, while first-order information guides block and pair selection. We establish the global convergence of BCD-SubG using a constrained Lojasiewicz inequality; additional eventual-interiority conditions yield stationary limits and convergence rates. We then specialize BCD-SubG to…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-12","reactions":{}}