Topology-Aware Dynamic Reweighting for Distribution Shifts on Graph

TL;DR

This paper introduces TAR, a topology-aware dynamic reweighting framework that improves graph neural network robustness against distribution shifts by leveraging graph structure and Wasserstein geometry, with proven theoretical guarantees.

Contribution

We propose TAR, a novel dynamic reweighting method that enhances OOD generalization in GNNs without relying on invariance assumptions, supported by theoretical proofs.

Findings

Significant performance improvements on four graph OOD datasets.

Effective handling of class imbalance in node classification.

Theoretical proof of distributional robustness.

Abstract

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.