FIT-GNN: Faster Inference Time for GNNs that 'FIT' in Memory Using Coarsening

TL;DR

This paper introduces a novel graph coarsening approach to significantly accelerate inference time and reduce memory usage in GNNs, enabling efficient deployment on low-resource devices.

Contribution

It proposes two methods, Extra Nodes and Cluster Nodes, extending graph coarsening to inference for graph-level tasks with extensive experimental validation.

Findings

Orders of magnitude faster inference on benchmark datasets.

Significant memory reduction for node and graph tasks.

Maintains competitive accuracy with baseline models.

Abstract

Scalability of Graph Neural Networks (GNNs) remains a significant challenge. To tackle this, methods like coarsening, condensation, and computation trees are used to train on a smaller graph, resulting in faster computation. Nonetheless, prior research has not adequately addressed the computational costs during the inference phase. This paper presents a novel approach to improve the scalability of GNNs by reducing computational burden during the inference phase using graph coarsening. We demonstrate two different methods -- Extra Nodes and Cluster Nodes. Our study extends the application of graph coarsening for graph-level tasks, including graph classification and graph regression. We conduct extensive experiments on multiple benchmark datasets to evaluate the performance of our approach. Our results show that the proposed method achieves orders of magnitude improvements in single-node inference time compared to traditional approaches. Furthermore, it significantly reduces memory consumption for node and graph classification and regression tasks, enabling efficient training and inference on low-resource devices where conventional methods are impractical. Notably, these computational advantages are achieved while maintaining competitive performance relative to baseline models.