Spectral Modification of Graphs for Improved Spectral Clustering

Part of Advances in Neural Information Processing Systems 32 (NeurIPS 2019)

AuthorFeedback Bibtex MetaReview Metadata Paper Reviews Supplemental


Ioannis Koutis, Huong Le


Spectral clustering algorithms provide approximate solutions to hard optimization problems that formulate graph partitioning in terms of the graph conductance. It is well understood that the quality of these approximate solutions is negatively affected by a possibly significant gap between the conductance and the second eigenvalue of the graph. In this paper we show that for \textbf{any} graph $G$, there exists a `spectral maximizer' graph $H$ which is cut-similar to $G$, but has eigenvalues that are near the theoretical limit implied by the cut structure of $G$. Applying then spectral clustering on $H$ has the potential to produce improved cuts that also exist in $G$ due to the cut similarity. This leads to the second contribution of this work: we describe a practical spectral modification algorithm that raises the eigenvalues of the input graph, while preserving its cuts. Combined with spectral clustering on the modified graph, this yields demonstrably improved cuts.