Normalize Then Propagate: Efficient Homophilous Regularization for Few-shot Semi-Supervised Node Classification

TL;DR

This paper introduces NormProp, a novel GNN regularization method that leverages homophily to improve semi-supervised node classification in low-label scenarios, achieving state-of-the-art results efficiently.

Contribution

The paper proposes NormProp, a new regularization technique that decouples node representation components and uses homophily to enhance GNN generalization with limited labels.

Findings

NormProp outperforms existing methods in low-label semi-supervised node classification.

It achieves state-of-the-art accuracy with low computational complexity.

Theoretical analysis provides bounds on node representation norms.

Abstract

Graph Neural Networks (GNNs) have demonstrated remarkable ability in semi-supervised node classification. However, most existing GNNs rely heavily on a large amount of labeled data for training, which is labor-intensive and requires extensive domain knowledge. In this paper, we first analyze the restrictions of GNNs generalization from the perspective of supervision signals in the context of few-shot semi-supervised node classification. To address these challenges, we propose a novel algorithm named NormProp, which utilizes the homophily assumption of unlabeled nodes to generate additional supervision signals, thereby enhancing the generalization against label scarcity. The key idea is to efficiently capture both the class information and the consistency of aggregation during message passing, via decoupling the direction and Euclidean norm of node representations. Moreover, we conduct a theoretical analysis to determine the upper bound of Euclidean norm, and then propose homophilous regularization to constraint the consistency of unlabeled nodes. Extensive experiments demonstrate that NormProp achieve state-of-the-art performance under low-label rate scenarios with low computational complexity.