Clique pooling for graph classification

TL;DR

This paper introduces a topological graph pooling method based on cliques, enhancing interpretability and compatibility with existing architectures, and demonstrates its effectiveness on standard benchmarks and regular graphs.

Contribution

A novel clique-based pooling operation for graphs that is purely topological, interpretable, nonparametric, and compatible with existing GCN and GraphSAGE models.

Findings

Competitive performance on graph classification benchmarks

Effective pooling in regular CNNs when replacing traditional pooling

Enhanced interpretability due to topological nature

Abstract

We propose a novel graph pooling operation using cliques as the unit pool. As this approach is purely topological, rather than featural, it is more readily interpretable, a better analogue to image coarsening than filtering or pruning techniques, and entirely nonparametric. The operation is implemented within graph convolution network (GCN) and GraphSAGE architectures and tested against standard graph classification benchmarks. In addition, we explore the backwards compatibility of the pooling to regular graphs, demonstrating competitive performance when replacing two-by-two pooling in standard convolutional neural networks (CNNs) with our mechanism.