Edge but not Least: Cross-View Graph Pooling

TL;DR

This paper introduces Co-Pooling, a cross-view graph pooling method that combines node and edge views to enhance graph-level representations, outperforming existing methods on multiple benchmarks.

Contribution

The paper proposes a novel cross-view pooling approach that integrates node and edge information for improved graph representation learning.

Findings

Outperforms state-of-the-art pooling methods on 15 benchmark datasets.

Effectively handles graphs with diverse node attributes.

Enhances graph classification and regression performance.

Abstract

Graph neural networks have emerged as a powerful model for graph representation learning to undertake graph-level prediction tasks. Various graph pooling methods have been developed to coarsen an input graph into a succinct graph-level representation through aggregating node embeddings obtained via graph convolution. However, most graph pooling methods are heavily node-centric and are unable to fully leverage the crucial information contained in global graph structure. This paper presents a cross-view graph pooling (Co-Pooling) method to better exploit crucial graph structure information. The proposed Co-Pooling fuses pooled representations learnt from both node view and edge view. Through cross-view interaction, edge-view pooling and node-view pooling seamlessly reinforce each other to learn more informative graph-level representations. Co-Pooling has the advantage of handling various graphs with different types of node attributes. Extensive experiments on a total of 15 graph benchmark datasets validate the effectiveness of our proposed method, demonstrating its superior performance over state-of-the-art pooling methods on both graph classification and graph regression tasks.