On the Robustness of Spectral Algorithms for Semirandom Stochastic Block Models

TL;DR

This paper investigates the robustness of spectral clustering algorithms for stochastic block models under semirandom adversaries, revealing conditions where unnormalized Laplacian methods succeed or fail in exact recovery.

Contribution

It identifies classes of semirandom adversaries where unnormalized spectral clustering is provably consistent and highlights limitations of normalized spectral methods under similar conditions.

Findings

Unnormalized Laplacian spectral clustering can exactly recover communities under certain semirandom adversaries.

Normalized Laplacian spectral clustering fails to achieve exact recovery in the same adversarial settings.

Numerical experiments support the theoretical robustness and limitations identified.

Abstract

In a graph bisection problem, we are given a graph $G$ with two equally-sized unlabeled communities, and the goal is to recover the vertices in these communities. A popular heuristic, known as spectral clustering, is to output an estimated community assignment based on the eigenvector corresponding to the second smallest eigenvalue of the Laplacian of $G$. Spectral algorithms can be shown to provably recover the cluster structure for graphs generated from certain probabilistic models, such as the Stochastic Block Model (SBM). However, spectral clustering is known to be non-robust to model mis-specification. Techniques based on semidefinite programming have been shown to be more robust, but they incur significant computational overheads. In this work, we study the robustness of spectral algorithms against semirandom adversaries. Informally, a semirandom adversary is allowed to ``helpfully'' change the specification of the model in a way that is consistent with the ground-truth solution. Our semirandom adversaries in particular are allowed to add edges inside clusters or increase the probability that an edge appears inside a cluster. Semirandom adversaries are a useful tool to determine the extent to which an algorithm has overfit to statistical assumptions on the input. On the positive side, we identify classes of semirandom adversaries under which spectral bisection using the _unnormalized_ Laplacian is strongly consistent, i.e., it exactly recovers the planted partitioning. On the negative side, we show that in these classes spectral bisection with the _normalized_ Laplacian outputs a partitioning that makes a classification mistake on a constant fraction of the vertices. Finally, we demonstrate numerical experiments that complement our theoretical findings.