Graph Domain Adaptation via Homophily-Agnostic Reconstructing Structure

2026-02-07

AI Summary

This work addresses a critical limitation in existing graph domain adaptation methods, which typically assume homophily in graph structures and consequently suffer significant performance degradation when the target graph exhibits unknown or heterophilous characteristics. To overcome this, we propose an unsupervised graph domain adaptation approach that does not rely on any homophily assumption. Our method constructs both highly homophilous and highly heterophilous variants of the source and target graphs and performs knowledge alignment between corresponding variants, thereby enabling robust transfer across arbitrary homophilous or heterophilous structures. This is the first domain adaptation framework designed to be agnostic to graph homophily. Extensive experiments on five benchmark datasets demonstrate that our method substantially outperforms state-of-the-art approaches, with particularly pronounced gains on heterophilous graphs, confirming its effectiveness and generalization capability.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Application Domains: Humanities & Computational Social Science

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

Graph Domain Adaptation (GDA) transfers knowledge from labeled source graphs to unlabeled target graphs, addressing the challenge of label scarcity. However, existing GDA methods typically assume that both source and target graphs exhibit homophily, leading existing methods to perform poorly when heterophily is present. Furthermore, the lack of labels in the target graph makes it impossible to assess its homophily level beforehand. To address this challenge, we propose a novel homophily-agnostic approach that effectively transfers knowledge between graphs with varying degrees of homophily. Specifically, we adopt a divide-and-conquer strategy that first separately reconstructs highly homophilic and heterophilic variants of both the source and target graphs, and then performs knowledge alignment separately between corresponding graph variants. Extensive experiments conducted on five benchmark datasets demonstrate the superior performance of our approach, particularly highlighting its substantial advantages on heterophilic graphs.

Problem

• Graph Domain Adaptation
• Homophily
• Heterophily
• Label Scarcity
• Knowledge Transfer

Innovation

• Graph Domain Adaptation
• Homophily-Agnostic
• Heterophily
• Structure Reconstruction
• Knowledge Alignment