G-Signatures: Global Graph Propagation With Randomized Signatures

TL;DR

G-Signatures is a new graph neural network method that enables global graph information processing through randomized signatures and latent space path traversal, improving scalability and performance on large graph tasks.

Contribution

It introduces a novel global graph propagation technique using randomized signatures and latent space path mapping, addressing over-smoothing in GNNs.

Findings

Outperforms existing methods on classification tasks

Effective in processing large-scale graphs

Enhances global property extraction

Abstract

Graph neural networks (GNNs) have evolved into one of the most popular deep learning architectures. However, GNNs suffer from over-smoothing node information and, therefore, struggle to solve tasks where global graph properties are relevant. We introduce G-Signatures, a novel graph learning method that enables global graph propagation via randomized signatures. G-Signatures use a new graph conversion concept to embed graph structured information which can be interpreted as paths in latent space. We further introduce the idea of latent space path mapping. This allows us to iteratively traverse latent space paths, and, thus globally process information. G-Signatures excel at extracting and processing global graph properties, and effectively scale to large graph problems. Empirically, we confirm the advantages of G-Signatures at several classification and regression tasks.