Papers › Signed Graph Metric Learning via Gershgorin Disc Perfect Alignment

Signed Graph Metric Learning via Gershgorin Disc Perfect Alignment

15 Jun 2020arXiv:2006.08816archive 2025-07-28

Cheng Yang, Gene Cheung, Wei Hu

Given a convex and differentiable objective Q() for a real symmetric matrix $\M$ in the positive definite (PD) cone -- used to compute Mahalanobis distances -- we propose a fast general metric learning framework that is entirely projection-free. We first assume that $\M$ resides in a space $\cS$ of generalized graph Laplacian matrices corresponding to balanced signed graphs. ∈ that is also PD is called a graph metric matrix. Unlike low-rank metric matrices common in the literature, $\cS$ includes the important diagonal-only matrices as a special case. The key theorem to circumvent full eigen-decomposition and enable fast metric matrix optimization is Gershgorin disc perfect alignment (GDPA): given ∈ and diagonal matrix $\S$, where Sᵢᵢ = 1/vᵢ and ̌ is $\M$'s first eigenvector, we prove that Gershgorin disc left-ends of similarity transform = ⁻¹ are perfectly aligned at the smallest eigenvalue λₘᵢₙ. Using this theorem, we replace the PD cone constraint in the metric learning problem with tightest possible linear constraints per iteration, so that the alternating optimization of the diagonal / off-diagonal terms in $\M$ can be solved efficiently as linear programs via the Frank-Wolfe method. We update ̌ using Locally Optimal Block Preconditioned Conjugate Gradient (LOBPCG) with warm start as entries in $\M$ are optimized successively. Experiments show that our graph metric optimization is significantly faster than cone-projection schemes, and produces competitive binary classification performance.

PaperPDFCode

Code

bobchengyang/SGML officialmentioned in paper report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Binary ClassificationMetric Learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections