Subgraph Neural Networks

TL;DR

SubGNN introduces a novel subgraph neural network with a routing mechanism and multiple channels to effectively learn subgraph representations, significantly improving subgraph classification accuracy on synthetic and real-world datasets.

Contribution

The paper presents SubGNN, a new model with a subgraph routing mechanism and multiple channels, addressing the challenge of subgraph prediction in graph neural networks.

Findings

SubGNN outperforms baseline methods by 19.8% on eight datasets.

It achieves high accuracy on complex biomedical subgraphs.

Empirical results validate the effectiveness of the proposed channels.

Abstract

Deep learning methods for graphs achieve remarkable performance on many node-level and graph-level prediction tasks. However, despite the proliferation of the methods and their success, prevailing Graph Neural Networks (GNNs) neglect subgraphs, rendering subgraph prediction tasks challenging to tackle in many impactful applications. Further, subgraph prediction tasks present several unique challenges: subgraphs can have non-trivial internal topology, but also carry a notion of position and external connectivity information relative to the underlying graph in which they exist. Here, we introduce SubGNN, a subgraph neural network to learn disentangled subgraph representations. We propose a novel subgraph routing mechanism that propagates neural messages between the subgraph's components and randomly sampled anchor patches from the underlying graph, yielding highly accurate subgraph representations. SubGNN specifies three channels, each designed to capture a distinct aspect of subgraph topology, and we provide empirical evidence that the channels encode their intended properties. We design a series of new synthetic and real-world subgraph datasets. Empirical results for subgraph classification on eight datasets show that SubGNN achieves considerable performance gains, outperforming strong baseline methods, including node-level and graph-level GNNs, by 19.8% over the strongest baseline. SubGNN performs exceptionally well on challenging biomedical datasets where subgraphs have complex topology and even comprise multiple disconnected components.