Beyond Fixed Depth: Adaptive Graph Neural Networks for Node Classification Under Varying Homophily

TL;DR

This paper introduces an adaptive-depth GNN framework that dynamically adjusts node-specific aggregation depths based on local structural and label information, improving node classification across both homophilic and heterophilic graphs.

Contribution

We develop a theoretical framework linking local graph properties to optimal information propagation depths and propose a novel adaptive-depth GNN architecture that generalizes across different homophily regimes.

Findings

Improved node classification accuracy on diverse benchmarks.

Adaptive-depth GNN outperforms fixed-depth models.

Theoretically grounded metrics effectively guide depth selection.

Abstract

Graph Neural Networks (GNNs) have achieved significant success in addressing node classification tasks. However, the effectiveness of traditional GNNs degrades on heterophilic graphs, where connected nodes often belong to different labels or properties. While recent work has introduced mechanisms to improve GNN performance under heterophily, certain key limitations still exist. Most existing models apply a fixed aggregation depth across all nodes, overlooking the fact that nodes may require different propagation depths based on their local homophily levels and neighborhood structures. Moreover, many methods are tailored to either homophilic or heterophilic settings, lacking the flexibility to generalize across both regimes. To address these challenges, we develop a theoretical framework that links local structural and label characteristics to information propagation dynamics at the node level. Our analysis shows that optimal aggregation depth varies across nodes and is critical for preserving class-discriminative information. Guided by this insight, we propose a novel adaptive-depth GNN architecture that dynamically selects node-specific aggregation depths using theoretically grounded metrics. Our method seamlessly adapts to both homophilic and heterophilic patterns within a unified model. Extensive experiments demonstrate that our approach consistently enhances the performance of standard GNN backbones across diverse benchmarks.