该论文首次定义了跨网络同嗜边和异嗜边分类 CNHHEC 问题。针对该问题,论文提出域自适应的图注意力监督网络模型 DGASN。DGASN 采用多头 GAT 构建 GNN 编码器,并将节点嵌入和边嵌入端到端地联合训练,从而学习信息丰富的嵌入以区分同嗜边和异嗜边。
为了提高原始 GAT 中图注意力的表达能力,DGASN 设计了直接监督的图注意力损失,基于源网络的已知边标签,对图注意力权重的学习进行直接监督,从而在邻居聚合过程中,为异嗜边赋予较低的注意力权重,而为同嗜边赋予较高的注意力权重。这将产生更具类别鉴别性的节点嵌入来区分不同类别的节点,从而得到更具类别鉴别性的边嵌入以区分同嗜边和异嗜边。
[1] X. Shen, M. Shao, S. Pan, L. T. Yang, and X. Zhou, 'Domain-adaptive Graph Attention-supervised Network for Cross-network Edge Classification,' IEEE Transactions on Neural Networks and Learning Systems, 2023.[2] C. Liu, J. Wu, W. Liu, and W. Hu, 'Enhancing graph neural networks by a high-quality aggregation of beneficial information,' Neural Networks, vol. 142, pp. 20-33, 2021.[3] O. Stretcu, K. Viswanathan, D. Movshovitz-Attias, E. Platanios, S. Ravi, and A. Tomkins, 'Graph Agreement Models for Semi-supervised Learning,' in Proc. NeurIPS, 2019, pp. 8713-8723.[4] H. Chen, Y. Xu, F. Huang, Z. Deng, W. Huang, S. Wang, P. He, and Z. Li, 'Label-aware Graph Convolutional Networks,' in Proc. CIKM, 2020, pp. 1977-1980.[5] E. Dai, W. Jin, H. Liu, and S. Wang, 'Towards Robust Graph Neural Networks for Noisy Graphs with Sparse Labels,' in Proc. WSDM, 2022, pp. 181-191.[6] E. Dai, C. Aggarwal, and S. Wang, 'Nrgnn: Learning a Label Noise Resistant Graph Neural Network on Sparsely And Noisily Labeled Graphs,' in Proc. SIGKDD, 2021, pp. 227-236.[7] H. Yang, X. Yan, X. Dai, Y. Chen, and J. Cheng, 'Self-enhanced GNN: Improving Graph Neural Networks Using Model Outputs,' in Proc. IJCNN, 2021, pp. 1-8.[8] P. Veličković, G. Cucurull, A. Casanova, A. Romero, P. Liò, and Y. Bengio, 'Graph Attention Networks,' in Proc. ICLR, 2018.[9] G. Wang, R. Ying, J. Huang, and J. Leskovec, 'Improving graph attention networks with large margin-based constraints,' in Proc. NeurIPS, 2019.[10] D. Kim and A. Oh, 'How to Find Your Friendly Neighborhood: Graph Attention Design with Self-supervision,' in Proc. ICLR, 2021.[11] X. Shen, Q. Dai, S. Mao, F.-l. Chung, and K.-S. Choi, 'Network Together: Node Classification via Cross network Deep Network Embedding,' IEEE Trans. Neural. Netw. Learn. Syst., vol. 32, no. 5, pp. 1935-1948, 2021.[12] T. N. Kipf and M. Welling, 'Variational Graph Auto-encoders,' in Proc. NeurIPS Workshop, 2016.[13] G. Cui, J. Zhou, C. Yang, and Z. Liu, 'Adaptive Graph Encoder for Attributed Graph Embedding,' in Proc. SIGKDD, 2020, pp. 976-985.[14] X. Shen, Q. Dai, F.-l. Chung, W. Lu, and K.-S. Choi, 'Adversarial Deep Network Embedding for Cross-network Node Classification,' in Proc. AAAI, 2020, pp. 2991-2999.[15] Q. Dai, X.-M. Wu, J. Xiao, X. Shen, and D. Wang, 'Graph Transfer Learning via Adversarial Domain Adaptation with Graph Convolution,' IEEE Trans. Knowl. Data Eng., 2022.[16] M. Wu, S. Pan, C. Zhou, X. Chang, and X. Zhu, 'Unsupervised Domain Adaptive Graph Convolutional Networks,' in Proc. WWW, 2020, pp. 1457-1467.[17] X. Zhang, Y. Du, R. Xie, and C. Wang, 'Adversarial Separation Network for Cross-Network Node Classification,' in Proc. CIKM, 2021, pp. 2618-2626.