Unperturbed: spectral analysis beyond Davis-Kahan

Abstract

Classical matrix perturbation results, such as Weyl's theorem for eigenvalues and the Davis-Kahan theorem for eigenvectors, are general purpose. These classical bounds are tight in the worst case, but in many settings sub-optimal in the typical case. In this paper, we present perturbation bounds which consider the nature of the perturbation and its interaction with the unperturbed structure in order to obtain significant improvements over the classical theory in many scenarios, such as when the perturbation is random. We demonstrate the utility of these new results by analyzing perturbations in the stochastic blockmodel where we derive much tighter bounds than provided by the classical theory. We use our new perturbation theory to show that a very simple and natural clustering algorithm -- whose analysis was difficult using the classical tools -- nevertheless recovers the communities of the blockmodel exactly even in very sparse graphs.

References in corpus (2)

• Angles between subspaces and their tangents
• Random perturbation of low rank matrices: Improving classical bounds

Cited by in corpus (10)

• Spectral Method and Regularized MLE Are Both Optimal for Top-$K$ Ranking
• Analysis of centrality measures under differential privacy models
• Unified ${\ell }_{2\to \infty }$ Eigenspace Perturbation Theory for Symmetric Random Matrices
• Eigen-convergence of Gaussian kernelized graph Laplacian by manifold heat interpolation
• Spectral Convergence of Graph Laplacian and Heat Kernel Reconstruction in $L^{\infty }$ from Random Samples
• Consistency of Spectral Clustering on Hierarchical Stochastic Block Models
• A Sharp Blockwise Tensor Perturbation Bound for Orthogonal Iteration
• Mask-GVAE: Blind Denoising Graphs via Partition
• Convergence of Graph Laplacian with kNN Self-tuned Kernels
• Contribution of directedness in graph spectra