Complete the Missing Half: Augmenting Aggregation Filtering with Diversification for Graph Convolutional Neural Networks

TL;DR

This paper introduces a diversification technique to augment traditional aggregation in GNNs, enhancing node distinction and improving performance across multiple node classification tasks.

Contribution

It proposes a dual filtering approach combining aggregation and diversification, which can be integrated into existing GNNs to prevent node representation collapse.

Findings

Significant performance improvements on 9 node classification datasets.

The dual filtering approach enriches node representations effectively.

The method is compatible with various GNN training strategies.

Abstract

The core operation of current Graph Neural Networks (GNNs) is the aggregation enabled by the graph Laplacian or message passing, which filters the neighborhood information of nodes. Though effective for various tasks, in this paper, we show that they are potentially a problematic factor underlying all GNN models for learning on certain datasets, as they force the node representations similar, making the nodes gradually lose their identity and become indistinguishable. Hence, we augment the aggregation operations with their dual, i.e. diversification operators that make the node more distinct and preserve the identity. Such augmentation replaces the aggregation with a two-channel filtering process that, in theory, is beneficial for enriching the node representations. In practice, the proposed two-channel filters can be easily patched on existing GNN methods with diverse training strategies, including spectral and spatial (message passing) methods. In the experiments, we observe desired characteristics of the models and significant performance boost upon the baselines on 9 node classification tasks.