Bridging Domain Adaptation and Graph Neural Networks: A Tensor-Based Framework for Effective Label Propagation

TL;DR

This paper introduces LP-TGNN, a tensor-based framework that combines graph neural networks with domain adaptation techniques to improve label transfer across different graph domains, achieving superior performance on real-world benchmarks.

Contribution

The paper presents a novel tensor-based GNN framework that integrates domain adaptation and label propagation, enhancing transferability across graph domains.

Findings

LP-TGNN outperforms baseline methods on multiple benchmarks.

The framework effectively reduces domain discrepancy.

Ablation studies confirm the importance of each component.

Abstract

Graph Neural Networks (GNNs) have recently become the predominant tools for studying graph data. Despite state-of-the-art performance on graph classification tasks, GNNs are overwhelmingly trained in a single domain under supervision, thus necessitating a prohibitively high demand for labels and resulting in poorly transferable representations. To address this challenge, we propose the Label-Propagation Tensor Graph Neural Network (LP-TGNN) framework to bridge the gap between graph data and traditional domain adaptation methods. It extracts graph topological information holistically with a tensor architecture and then reduces domain discrepancy through label propagation. It is readily compatible with general GNNs and domain adaptation techniques with minimal adjustment through pseudo-labeling. Experiments on various real-world benchmarks show that our LP-TGNN outperforms baselines by a notable margin. We also validate and analyze each component of the proposed framework in the ablation study.