PALA: Class-imbalanced Graph Domain Adaptation via Prototype-anchored Learning and Alignment
Lists
Only you see this
Abstract
Graph domain adaptation is a key subfield of graph transfer learning that aims to bridge domain gaps by transferring knowledge from a label-rich source graph to an unlabeled target graph. However, most existing methods assume balanced labels in the source graph, which often fails in practice and leads to biased knowledge transfer. To address this, in this paper, we propose a prototype-anchored learning and alignment framework for class-imbalanced graph domain adaptation. Specifically, we incorporate pointwise node mutual information into the graph encoder to capture high-order topological proximity and learn generalized node representations. Leveraging this, we then introduce categorical prototypes with adversarial proto-instances for prototype-anchored learning and recalibration to represent the source graph under an imbalanced class distribution. Finally, we introduce a weighted prototype contrastive adaptation strategy that aligns target pseudo-labels with source prototypes to handle class imbalance during adaptation. Extensive experiments show that our PALA outperforms the state-of-the-art methods. Our code is available at https://github.com/maxin88scu/PALA.
BibTeX
@inproceedings{ma2025pala,
author = {Xin Ma and Yifan Wang and Siyu Yi and Wei Ju and Bei Wu and Ziyue Qiao and Chenwei Tang and Jiancheng Lv},
title = {{PALA: Class-imbalanced Graph Domain Adaptation via Prototype-anchored Learning and Alignment}},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2025},
doi = {10.24963/ijcai.2025/356},
}Loading related content...