Adversarial Contrastive Graph Masked AutoEncoder Against Graph Structure and Feature Dual Attacks

Graph Neural Networks (GNNs) have been shown vulnerable to graph adversarial attacks. Current robust graph representation learning methods mainly defend against graph structure attack, and improve performance of GNNs. However, node features in graph can also be easily attacked in reality. The joint defense on graph structure and feature dual attacks remains challenging yet less studied. To fulfill this gap, we propose Adversarial Contrastive Graph Masked AutoEncoder (ACGMAE) to defend against graph structure and feature dual attacks. ACGMAE employs adversarial feature masking for reconstructing node features to mitigate the influence of feature attack. Additionally, ACGMAE employs contrastive learning on kNN graph and attacked graph, considering neighbor nodes as positive samples. By calculating the probabilities of these neighbors being true positive, ACG-MAE effectively reduces the influence of adversarial edges. Extensive experiments on node classification and clustering tasks demonstrate the effectiveness of the proposed ACG-MAE, especially under graph structure and feature dual attacks.