Optimization flows landing on the Stiefel manifold⋆. Issue 30 (2022)

• Record Type: Journal Article
• Authors: Gao, Bin; Vary, Simon; Ablin, Pierre; Absil, P.-A.
• Is Part Of: IFAC-PapersOnLine. Volume 55:Issue 30(2022)
• Page Start: 25
• Page End: 30
• Publication Date: 2022
• Subjects: Stiefel manifold -- Landing flow -- Canonical metric -- Riemannian gradient -- Asymptotic stability; 37N40 -- 90C48; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2022.11.023
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: We study a continuous-time system that solves optimization problems over the set of orthonormal matrices, which is also known as the Stiefel manifold. The resulting optimization flow follows a path that is not always on the manifold but asymptotically lands on the manifold. We introduce a generalized Stiefel manifold to which we extend the canonical metric of the Stiefel manifold. We show that the vector field of the proposed flow can be interpreted as the sum of a Riemannian gradient on a generalized Stiefel manifold and a normal vector. Moreover, we prove that the proposed flow globally converges to the set of critical points, and any local minimum and isolated critical point is asymptotically stable.