Sparse Decomposition of Graph Neural Networks

TL;DR

This paper introduces a sparse decomposition method for GNNs that reduces inference costs by approximating node representations with a subset of neighbors, maintaining accuracy while improving efficiency.

Contribution

It proposes a novel sparse decomposition approach with an algorithm for optimal parameter computation, enabling faster inference with minimal accuracy loss.

Findings

Achieves linear complexity with respect to node degree and layers.

Outperforms baselines in inference speedup and accuracy.

Effective for node classification and spatio-temporal forecasting.

Abstract

Graph Neural Networks (GNN) exhibit superior performance in graph representation learning, but their inference cost can be high, due to an aggregation operation that can require a memory fetch for a very large number of nodes. This inference cost is the major obstacle to deploying GNN models with \emph{online prediction} to reflect the potentially dynamic node features. To address this, we propose an approach to reduce the number of nodes that are included during aggregation. We achieve this through a sparse decomposition, learning to approximate node representations using a weighted sum of linearly transformed features of a carefully selected subset of nodes within the extended neighbourhood. The approach achieves linear complexity with respect to the average node degree and the number of layers in the graph neural network. We introduce an algorithm to compute the optimal parameters for the sparse decomposition, ensuring an accurate approximation of the original GNN model, and present effective strategies to reduce the training time and improve the learning process. We demonstrate via extensive experiments that our method outperforms other baselines designed for inference speedup, achieving significant accuracy gains with comparable inference times for both node classification and spatio-temporal forecasting tasks.