Distribution Preserving Graph Representation Learning

TL;DR

This paper introduces DP-GNN, a graph neural network framework that enhances generalization by preserving distribution information in representations, achieving state-of-the-art results on graph classification benchmarks.

Contribution

The paper proposes a novel DP-GNN framework that balances expressive power and generalization by preserving distribution information in graph and node representations.

Findings

DP-GNN achieves state-of-the-art performance on multiple benchmarks.

Preserving distribution information improves GNN generalization.

The framework maintains high expressive power while enhancing robustness.

Abstract

Graph neural network (GNN) is effective to model graphs for distributed representations of nodes and an entire graph. Recently, research on the expressive power of GNN attracted growing attention. A highly-expressive GNN has the ability to generate discriminative graph representations. However, in the end-to-end training process for a certain graph learning task, a highly-expressive GNN risks generating graph representations overfitting the training data for the target task, while losing information important for the model generalization. In this paper, we propose Distribution Preserving GNN (DP-GNN) - a GNN framework that can improve the generalizability of expressive GNN models by preserving several kinds of distribution information in graph representations and node representations. Besides the generalizability, by applying an expressive GNN backbone, DP-GNN can also have high expressive power. We evaluate the proposed DP-GNN framework on multiple benchmark datasets for graph classification tasks. The experimental results demonstrate that our model achieves state-of-the-art performances.