Split-Merge: A Difference-based Approach for Dominant Eigenvalue Problem

Abstract

The computation of the dominant eigenpair for symmetric positive semidefinite matrices is fundamental in numerical optimization. This work shifts the paradigm from the classical Rayleigh quotient to an unconstrained difference formulation, whose global optimum recovers the dominant eigenpair. Within this framework, we prove that gradient descent with a constant step-size $\alpha \in (0, 1)$ converges almost surely to the global optimum at a local linear rate. This analysis thereby reinterprets the classical power method as the conservative special case $\alpha=1/2$ and rigorously establishes its asymptotic sub-optimality. To advance this first-order scheme, we propose the Split-Merge algorithm based on the majorization-minimization principle. After splitting the matrix, we introduce auxiliary vectors to effectively merge the decomposition factors, resulting in a matrix-free and parameter-free iteration that captures tighter curvature information. We establish that Split-Merge converges almost surely to a global minimizer, and show that the iteration exhibits a spectral peeling mechanism that suppresses the targeted eigenspace, potentially surpassing the static linear rate of power iterations. Numerical evaluations across synthetic and real-world datasets confirm that our method has scalable efficiency, achieving speed-ups exceeding $10\times$ over the power method, with performance comparable to subspace iterations.