Coloring graph neural networks for node disambiguation

TL;DR

This paper introduces CLIP, a coloring-based graph neural network that enhances expressive power and node disambiguation, achieving state-of-the-art results in graph classification tasks.

Contribution

The paper proposes a simple coloring scheme for GNNs, providing a universal approximation capability and improved structural discrimination over traditional message passing neural networks.

Findings

CLIP is a universal approximator of functions on graphs.

CLIP outperforms traditional MPNNs on benchmark datasets.

Coloring improves the ability to distinguish structural graph features.

Abstract

In this paper, we show that a simple coloring scheme can improve, both theoretically and empirically, the expressive power of Message Passing Neural Networks(MPNNs). More specifically, we introduce a graph neural network called Colored Local Iterative Procedure (CLIP) that uses colors to disambiguate identical node attributes, and show that this representation is a universal approximator of continuous functions on graphs with node attributes. Our method relies on separability , a key topological characteristic that allows to extend well-chosen neural networks into universal representations. Finally, we show experimentally that CLIP is capable of capturing structural characteristics that traditional MPNNs fail to distinguish,while being state-of-the-art on benchmark graph classification datasets.