GC-Flow: A Graph-Based Flow Network for Effective Clustering

TL;DR

GC-Flow introduces a graph-based generative model that combines GCNs with normalizing flows, enabling effective clustering while maintaining classification performance on graph data.

Contribution

It replaces GCN layers with normalizing flows to create a generative model that improves clustering without sacrificing classification accuracy.

Findings

GC-Flow achieves well-separated clusters on benchmark datasets.

Incorporating adjacency matrix parameterization further improves clustering.

Maintains predictive power of GCN while enhancing clustering quality.

Abstract

Graph convolutional networks (GCNs) are \emph{discriminative models} that directly model the class posterior $p(y|\mathbf{x})$ for semi-supervised classification of graph data. While being effective, as a representation learning approach, the node representations extracted from a GCN often miss useful information for effective clustering, because the objectives are different. In this work, we design normalizing flows that replace GCN layers, leading to a \emph{generative model} that models both the class conditional likelihood $p(\mathbf{x}|y)$ and the class prior $p(y)$. The resulting neural network, GC-Flow, retains the graph convolution operations while being equipped with a Gaussian mixture representation space. It enjoys two benefits: it not only maintains the predictive power of GCN, but also produces well-separated clusters, due to the structuring of the representation space. We demonstrate these benefits on a variety of benchmark data sets. Moreover, we show that additional parameterization, such as that on the adjacency matrix used for graph convolutions, yields additional improvement in clustering.