Scalable Expressiveness through Preprocessed Graph Perturbations

TL;DR

This paper introduces SE2P, a scalable graph neural network method that balances expressiveness and efficiency through preprocessed graph perturbations, enabling faster training with improved generalizability.

Contribution

SE2P provides a flexible framework with configurable trade-offs between scalability and generalizability for graph neural networks.

Findings

SE2P achieves up to 8-fold speed improvements.

SE2P enhances generalizability over existing benchmarks.

SE2P's configurations effectively balance scalability and expressiveness.

Abstract

Graph Neural Networks (GNNs) have emerged as the predominant method for analyzing graph-structured data. However, canonical GNNs have limited expressive power and generalization capability, thus triggering the development of more expressive yet computationally intensive methods. One such approach is to create a series of perturbed versions of input graphs and then repeatedly conduct multiple message-passing operations on all variations during training. Despite their expressive power, this approach does not scale well on larger graphs. To address this scalability issue, we introduce Scalable Expressiveness through Preprocessed Graph Perturbation (SE2P). This model offers a flexible, configurable balance between scalability and generalizability with four distinct configuration classes. At one extreme, the configuration prioritizes scalability through minimal learnable feature extraction and extensive preprocessing; at the other extreme, it enhances generalizability with more learnable feature extractions, though this increases scalability costs. We conduct extensive experiments on real-world datasets to evaluate the generalizability and scalability of SE2P variants compared to various state-of-the-art benchmarks. Our results indicate that, depending on the chosen SE2P configuration, the model can enhance generalizability compared to benchmarks while achieving significant speed improvements of up to 8-fold.