Topology-Aware Dynamic Reweighting for Distribution Shifts on Graph
This addresses distribution shift problems in graph neural networks for node classification tasks, representing an incremental improvement over existing invariant learning approaches.
The paper tackles the problem of graph neural networks failing to generalize when training and test nodes come from different distributions by introducing the Topology-Aware Dynamic Reweighting (TAR) framework, which dynamically adjusts sample weights through gradient flow in geometric Wasserstein space and demonstrates marked improvements on four graph OOD datasets and three class-imbalanced datasets.
Graph Neural Networks (GNNs) are widely used for node classification tasks but often fail to generalize when training and test nodes come from different distributions, limiting their practicality. To overcome this, recent approaches adopt invariant learning techniques from the out-of-distribution (OOD) generalization field, which seek to establish stable prediction methods across environments. However, the applicability of these invariant assumptions to graph data remains unverified, and such methods often lack solid theoretical support. In this work, we introduce the Topology-Aware Dynamic Reweighting (TAR) framework, which dynamically adjusts sample weights through gradient flow in the geometric Wasserstein space during training. Instead of relying on strict invariance assumptions, we prove that our method is able to provide distributional robustness, thereby enhancing the out-of-distribution generalization performance on graph data. By leveraging the inherent graph structure, TAR effectively addresses distribution shifts. Our framework's superiority is demonstrated through standard testing on four graph OOD datasets and three class-imbalanced node classification datasets, exhibiting marked improvements over existing methods.