Refining Latent Homophilic Structures over Heterophilic Graphs for Robust Graph Convolution Networks
Chenyang Qiu, Guoshun Nan, Tianyu Xiong, Wendi Deng, Di Wang, Zhiyang Teng, Lijuan Sun, Qimei Cui
Abstract
Graph convolution networks (GCNs) are extensively utilized in various graph tasks to mine knowledge from spatial data. Our study marks the pioneering attempt to quantitatively investigate the GCN robustness over omnipresent heterophilic graphs for node classification. We uncover that the predominant vulnerability is caused by the structural out-of-distribution (OOD) issue. This finding motivates us to present a novel method that aims to harden GCNs by automatically learning Latent Homophilic Structures over heterophilic graphs. We term such a methodology as LHS. To elaborate, our initial step involves learning a latent structure by employing a novel self-expressive technique based on multi-node interactions. Subsequently, the structure is refined using a pairwisely constrained dual-view contrastive learning approach. We iteratively perform the above procedure, enabling a GCN model to aggregate information in a homophilic way on heterophilic graphs. Armed with such an adaptable structure, we can properly mitigate the structural OOD threats over heterophilic graphs. Experiments on various benchmarks show the effectiveness of the proposed LHS approach for robust GCNs.
BibTeX
@article{Qiu_Nan_Xiong_Deng_Wang_Teng_Sun_Cui_Tao_2024, title={Refining Latent Homophilic Structures over Heterophilic Graphs for Robust Graph Convolution Networks}, volume={38}, url={https://ojs.aaai.org/index.php/AAAI/article/view/28741}, DOI={10.1609/aaai.v38i8.28741}, abstractNote={Graph convolution networks (GCNs) are extensively utilized in various graph tasks to mine knowledge from spatial data. Our study marks the pioneering attempt to quantitatively investigate the GCN robustness over omnipresent heterophilic graphs for node classification. We uncover that the predominant vulnerability is caused by the structural out-of-distribution (OOD) issue. This finding motivates us to present a novel method that aims to harden GCNs by automatically learning Latent Homophilic Structures over heterophilic graphs. We term such a methodology as LHS. To elaborate, our initial step involves learning a latent structure by employing a novel self-expressive technique based on multi-node interactions. Subsequently, the structure is refined using a pairwisely constrained dual-view contrastive learning approach. We iteratively perform the above procedure, enabling a GCN model to aggregate information in a homophilic way on heterophilic graphs. Armed with such an adaptable structure, we can properly mitigate the structural OOD threats over heterophilic graphs. Experiments on various benchmarks show the effectiveness of the proposed LHS approach for robust GCNs.}, number={8}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Qiu, Chenyang and Nan, Guoshun and Xiong, Tianyu and Deng, Wendi and Wang, Di and Teng, Zhiyang and Sun, Lijuan and Cui, Qimei and Tao, Xiaofeng}, year={2024}, month={Mar.}, pages={8930-8938} }