InfraredGP: Efficient Graph Partitioning via Spectral Graph Neural Networks with Negative Corrections

TL;DR

InfraredGP introduces a novel spectral graph neural network approach with negative corrections that efficiently performs graph partitioning without training, leveraging low-frequency information beyond traditional ranges to improve community detection.

Contribution

It proposes a training-free spectral GNN method with negative correction for graph partitioning, achieving high efficiency and competitive accuracy on benchmark datasets.

Findings

Achieves 16x-23x faster partitioning than baselines.

Performs well on static and streaming graph partitioning tasks.

Utilizes negative correction to enhance low-frequency information for community detection.

Abstract

Graph partitioning (GP), a.k.a. community detection, is a classic problem that divides nodes of a graph into densely-connected blocks. From a perspective of graph signal processing, we find that graph Laplacian with a negative correction can derive graph frequencies beyond the conventional range $[0, 2]$. To explore whether the low-frequency information beyond this range can encode more informative properties about community structures, we propose InfraredGP. It (\romannumeral1) adopts a spectral GNN as its backbone combined with low-pass filters and a negative correction mechanism, (\romannumeral2) only feeds random inputs to this backbone, (\romannumeral3) derives graph embeddings via one feed-forward propagation (FFP) without any training, and (\romannumeral4) obtains feasible GP results by feeding the derived embeddings to BIRCH. Surprisingly, our experiments demonstrate that based solely on the negative correction mechanism that amplifies low-frequency information beyond $[0, 2]$, InfraredGP can derive distinguishable embeddings for some standard clustering modules (e.g., BIRCH) and obtain high-quality results for GP without any training. Following the IEEE HPEC Graph Challenge benchmark, we evaluate InfraredGP for both static and streaming GP, where InfraredGP can achieve much better efficiency (e.g., 16x-23x faster) and competitive quality over various baselines. We have made our code public at https://github.com/KuroginQin/InfraredGP