[论文解读] Anonymized GCN: A Novel Robust Graph Embedding Method via Hiding Node Position in Noise.

Ao Liu|arXiv (Cornell University)|May 6, 2020

Advanced Graph Neural Networks参考文献 71被引用 5

一句话总结

本文提出了一种名为 Anonymized GCN (AN-GCN) 的鲁棒图神经网络，该模型在对抗性训练过程中通过噪声隐藏节点连接信息，实现仅使用节点编号和特征进行节点预测。通过从潜在噪声生成图结构并训练生成器-判别器对，AN-GCN 在节点预测任务中实现了更高的鲁棒性，并将准确率提升了 0.0293。

ABSTRACT

Graph convolution network (GCN) have achieved state-of-the-art performance in the task of node prediction in the graph structure. However, with the gradual various of graph attack methods, there are lack of research on the robustness of GCN. At this paper, we will design a robust GCN method for node prediction tasks. Considering the graph structure contains two types of information: node information and connection information, and attackers usually modify the connection information to complete the interference with the prediction results of the node, we first proposed a method to hide the connection information in the generator, named Anonymized GCN (AN-GCN). By hiding the connection information in the graph structure in the generator through adversarial training, the accurate node prediction can be completed only by the node number rather than its specific position in the graph. Specifically, we first demonstrated the key to determine the embedding of a specific node: the row corresponding to the node of the eigenmatrix of the Laplace matrix, by target it as the output of the generator, we designed a method to hide the node number in the noise. Take the corresponding noise as input, we will obtain the connection structure of the node instead of directly obtaining. Then the encoder and decoder are spliced both in discriminator, so that after adversarial training, the generator and discriminator can cooperate to complete the encoding and decoding of the graph, then complete the node prediction. Finally, All node positions can generated by noise at the same time, that is to say, the generator will hides all the connection information of the graph structure. The evaluation shows that we only need to obtain the initial features and node numbers of the nodes to complete the node prediction, and the accuracy did not decrease, but increased by 0.0293.

研究动机与目标

• 为应对图结构攻击对 GCN 模型在节点预测任务中日益增长的威胁。
• 通过将节点预测与对结构位置信息的依赖解耦，提升 GCN 的鲁棒性。
• 设计一种方法，仅使用节点编号和特征即可实现准确的节点预测，而无需依赖图拓扑结构。
• 开发一种生成框架，将连接信息隐藏在潜在噪声中，从而增强模型的抗御能力。

提出的方法

• 生成器被训练为从随机噪声生成图连接结构，从而有效隐藏图中真实节点位置。
• 生成器的输出以节点编号和噪声为条件，其目标输出为拉普拉斯矩阵中对应于该节点的行。
• 判别器在编码器和解码器分支上联合训练，以区分真实与生成的图结构，从而实现对抗性训练。
• 通过仅编码节点特征和编号而非拓扑结构，生成器学习生成对结构位置不敏感的节点嵌入。
• 该框架利用对抗性训练，确保生成器仅从噪声中学习重建连接模式。
• 训练完成后，生成器可从噪声中生成所有节点位置，从而有效实现图结构的匿名化。

实验结果

研究问题

• RQ1 能否通过将节点预测与图中显式的结构位置信息解耦，使其更具鲁棒性？
• RQ2 生成模型能否在保留预测准确率的同时，有效将连接信息隐藏在潜在噪声中？
• RQ3 生成器与判别器之间的对抗性训练是否能提升对图结构攻击的鲁棒性？
• RQ4 能否仅使用节点编号和初始特征进行准确的节点预测，而无需访问原始图拓扑结构？

主要发现

• AN-GCN 实现了更高的节点预测准确率，相比基线 GCN 模型准确率提升了 0.0293。
• 即使攻击者修改图连接，模型仍能保持高性能，展现出显著增强的鲁棒性。
• 生成器成功学习从噪声生成图结构，有效实现了节点位置的匿名化。
• 仅使用节点编号和初始特征即可实现节点预测，无需依赖结构拓扑。
• 对抗性训练过程使生成器与判别器能够共同学习到鲁棒且可泛化的图表示。
• 该方法有效将连接信息隐藏在潜在空间中，降低了对结构图攻击的脆弱性。