Entrywise Eigenvector Analysis of Random Matrices with Low Expected Rank

Recovering low-rank structures via eigenvector perturbation analysis is a common problem in statistical machine learning, such as in factor analysis, community detection, ranking, matrix completion, among others. While a large variety of results provide tight bounds on the average errors between empirical and population statistics of eigenvectors, fewer results are tight for entrywise analyses, which are critical for a number of problems such as community detection and ranking. This paper investigates the entrywise perturbation analysis for a large class of random matrices whose expectations are low-rank, including community detection, synchronization ($\mathbb{Z}_2$-spiked Wigner model) and matrix completion models. Denoting by $\{u_k\}$, respectively $\{u_k^*\}$, the eigenvectors of a random matrix $A$, respectively $\mathbb{E} A$, the paper characterizes cases for which $$u_k \approx \frac{A u_k^*}{\lambda_k^*}$$ serves as a first-order approximation under the $\ell_\infty$ norm. The fact that the approximation is both tight and linear in the random matrix $A$ allows for sharp comparisons of $u_k$ and $u_k^*$. In particular, it allows to compare the signs of $u_k$ and $u_k^*$ even when $\| u_k - u_k^*\|_{\infty}$ is large, which in turn allows to settle the conjecture in Abbe et al. (2016) that the spectral algorithm achieves exact recovery in the stochastic block model without any trimming or cleaning steps. The results are further extended to the perturbation of eigenspaces, providing new bounds for $\ell_\infty$-type errors in noisy matrix completion.