{"id":"arxiv/2611.24101","url":"https://arxiv.org/abs/2611.24101","tags":["math.OC"],"score":0,"title":"A Euclidean Quasi-Newton Method for Matrix-Manifold Optimization via Almost-Global Coordinates","domain":"math.OC","source":"arxiv","authors":["Ziang Tian","Jiang Hu","Zhigang Yao","Bingjie Li"],"reviews":0,"subject":"math.OC","summary":"We study smooth matrix-manifold optimization using objective values and first derivatives. Almost-global coordinates transform the manifold-constrained problem into an unconstrained pullback problem, to which we apply BFGS with a chart-restart strategy and a limited-memory implementation. We establish correspondence of first- and second-order stationary points and minimizers, together with conditional convergence and local q-superlinear convergence. The local rate follows from Euclidean full-memory BFGS theory once the chart is fixed, without the locking condition required by classical…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-11-18","reactions":{}}