Revisiting, Benchmarking and Understanding Unsupervised Graph Domain Adaptation

TL;DR

The first comprehensive benchmark for unsupervised graph domain adaptation named GDABench is presented, which encompasses 16 algorithms across 5 datasets with 74 adaptation tasks, and it is found that with appropriate neighbourhood aggregation mechanisms, simple GNN variants can even surpass state-of-the-art UGDA baselines.

Abstract

Unsupervised Graph Domain Adaptation (UGDA) involves the transfer of knowledge from a label-rich source graph to an unlabeled target graph under domain discrepancies. Despite the proliferation of methods designed for this emerging task, the lack of standard experimental settings and fair performance comparisons makes it challenging to understand which and when models perform well across different scenarios. To fill this gap, we present the first comprehensive benchmark for unsupervised graph domain adaptation named GDABench, which encompasses 16 algorithms across 5 datasets with 74 adaptation tasks. Through extensive experiments, we observe that the performance of current UGDA models varies significantly across different datasets and adaptation scenarios. Specifically, we recognize that when the source and target graphs face significant distribution shifts, it is imperative to formulate strategies to effectively address and mitigate graph structural shifts. We also find that with appropriate neighbourhood aggregation mechanisms, simple GNN variants can even surpass state-of-the-art UGDA baselines. To facilitate reproducibility, we have developed an easy-to-use library PyGDA for training and evaluating existing UGDA methods, providing a standardized platform in this community. Our source codes and datasets can be found at: https://github.com/pygda-team/pygda.

Figures (8)

• Figure 1: This timeline illustrates the diverse UGDA algorithms revisited in this paper. All of them are incorporated into our PyGDA library. More details are shown in Section \ref{['Sec:related-work']} and Appendix \ref{['Sec:appendix-models']}
• Figure 2: The combination process of SimGDA / SimGDA+.
• Figure 3: SimGDA: the compared performance of vanilla DA with 6 GNN variants.
• Figure 4: Label distribution of GDABench datasets.
• Figure 5: The compared performance of vanilla DA with 6 GNN variants.