OOD-GNN: Out-of-Distribution Generalized Graph Neural Network

TL;DR

This paper introduces OOD-GNN, a novel graph neural network designed to maintain high performance on unseen graphs with different distributions by eliminating spurious correlations through a nonlinear decorrelation method.

Contribution

The paper proposes a new out-of-distribution generalization method for GNNs using nonlinear decorrelation with random Fourier features and a global weight estimator.

Findings

Outperforms state-of-the-art baselines on synthetic and real-world datasets.

Effectively reduces distribution shift impact on GNN performance.

Demonstrates significant improvements in out-of-distribution generalization.

Abstract

Graph neural networks (GNNs) have achieved impressive performance when testing and training graph data come from identical distribution. However, existing GNNs lack out-of-distribution generalization abilities so that their performance substantially degrades when there exist distribution shifts between testing and training graph data. To solve this problem, in this work, we propose an out-of-distribution generalized graph neural network (OOD-GNN) for achieving satisfactory performance on unseen testing graphs that have different distributions with training graphs. Our proposed OOD-GNN employs a novel nonlinear graph representation decorrelation method utilizing random Fourier features, which encourages the model to eliminate the statistical dependence between relevant and irrelevant graph representations through iteratively optimizing the sample graph weights and graph encoder. We further present a global weight estimator to learn weights for training graphs such that variables in graph representations are forced to be independent. The learned weights help the graph encoder to get rid of spurious correlations and, in turn, concentrate more on the true connection between learned discriminative graph representations and their ground-truth labels. We conduct extensive experiments to validate the out-of-distribution generalization abilities on two synthetic and 12 real-world datasets with distribution shifts. The results demonstrate that our proposed OOD-GNN significantly outperforms state-of-the-art baselines.