Graph neural networks (GNNs) have made rapid developments in the recent years. Due to their great ability in modeling graph-structured data, GNNs are vastly used in various applications, including high-stakes scenarios such as financial analysis, traffic predictions, and drug discovery. Despite their great potential in benefiting humans in the real world, recent study shows that GNNs can leak private information, are vulnerable to adversarial attacks, can inherit and magnify societal bias from training data and lack interpretability, which have risk of causing unintentional harm to the users and society. For example, existing works demonstrate that attackers can fool the GNNs to give the outcome they desire with unnoticeable perturbation on training graph. GNNs trained on social networks may embed the discrimination in their decision process, strengthening the undesirable societal bias. Consequently, trust-worthy GNNs in various aspects are emerging to prevent the harm from GNN models and increase the users’ trust in GNNs. In this paper, we give a comprehensive survey of GNNs in the computational aspects of privacy, robustness, fairness, and explainability. For each aspect, we give the taxonomy of the related methods and formulate the general frameworks for the multiple categories of trustworthy GNNs. We also discuss the future research directions of each aspect and connections between these aspects to help achieve trustworthiness.
Graph Neural Networks (GNNs) have achieved promising results in various tasks such as node classification and graph classification. Recent studies find that GNNs are vulnerable to adversarial attacks. However, effective backdoor attacks on graphs are still an open problem. In particular, backdoor attack poisons the graph by attaching triggers and the target class label to a set of nodes in the training graph. The backdoored GNNs trained on the poisoned graph will then be misled to predict test nodes to target class once attached with triggers. Though there are some initial efforts in graph backdoor attacks, our empirical analysis shows that they may require a large attack budget for effective backdoor attacks and the injected triggers can be easily detected and pruned. Therefore, in this paper, we study a novel problem of unnoticeable graph backdoor attacks with limited attack budget. To fully utilize the attack budget, we propose to deliberately select the nodes to inject triggers and target class labels in the poisoning phase. An adaptive trigger generator is deployed to obtain effective triggers that are difficult to be noticed. Extensive experiments on real-world datasets against various defense strategies demonstrate the effectiveness of our proposed method in conducting effective unnoticeable backdoor attacks.
22 August 2022 in IEEE Transactions on Knowledge and Data Engineering
Graph neural networks (GNNs) have shown great power in modeling graph structured data. However, similar to other machine learning models, GNNs may make biased predictions w.r.t protected sensitive attributes, e.g., skin color and gender. This is because machine learning algorithms including GNNs are trained to reflect the distribution of the training data which often contains historical bias towards sensitive attributes. In addition, we empirically show that the discrimination in GNNs can be magnified by graph structures and the message-passing mechanism of GNNs. As a result, the applications of GNNs in high-stake domains such as crime rate prediction would be largely limited. Though extensive studies of fair classification have been conducted on independently and identically distributed (i.i.d) data, methods to address the problem of discrimination on non-i.i.d data are rather limited. Generally, learning fair models require abundant sensitive attributes to regularize the model. However, for many graphs such as social networks, users are reluctant to share sensitive attributes. Thus, only limited sensitive attributes are available for fair GNN training in practice. Moreover, directly collecting and applying the sensitive attributes in fair model training may cause privacy issues, because the sensitive information can be leaked in data breach or attacks on the trained model. Therefore, we study a novel and important problem of learning fair GNNs with limited number of private sensitive attributes, i.e., sensitive attributes that are processed with a privacy-preserving mechanism. In an attempt to address these problems, FairGNN is proposed to eliminate the bias of GNNs whilst maintaining high node classification accuracy by leveraging graph structures and limited sensitive information. To further preserve the privacy, private sensitive attributes with privacy guarantee are obtained by injecting noise based on local differential privacy. And We further extend FairGNN to NT-FairGNN to handle the limited and private sensitive attributes to simultaneously achieve fairness and preserve privacy. Theoretical analysis and extensive experiments on real-world datasets demonstrate the effectiveness of FairGNN and NT-FairGNN in achieving fair and high-accurate classification.
Graph Neural Networks (GNNs) have shown remarkable performance in various tasks. However, recent works reveal that GNNs are vulnerable to backdoor attacks. Generally, backdoor attack poisons the graph by attaching backdoor triggers and the target class label to a set of nodes in the training graph. A GNN trained on the poisoned graph will then be misled to predict test nodes attached with trigger to the target class. Despite their effectiveness, our empirical analysis shows that triggers generated by existing methods tend to be out-of-distribution (OOD), which significantly differ from the clean data. Hence, these injected triggers can be easily detected and pruned with widely used outlier detection methods in real-world applications. Therefore, in this paper, we study a novel problem of unnoticeable graph backdoor attacks with in-distribution (ID) triggers. To generate ID triggers, we introduce an OOD detector in conjunction with an adversarial learning strategy to generate the attributes of the triggers within distribution. To ensure a high attack success rate with ID triggers, we introduce novel modules designed to enhance trigger memorization by the victim model trained on poisoned graph. Extensive experiments on real-world datasets demonstrate the effectiveness of the proposed method in generating in distribution triggers that can bypass various defense strategies while maintaining a high attack success rate.
14 February 2025 in ACM Transactions on Knowledge Discovery From Data
Graph Neural Networks (GNNs) have shown great ability in modeling graph-structured data for various domains. However, GNNs are known as black-box models that lack interpretability. Without understanding their inner working, we cannot fully trust them, which largely limits their adoption in high-stake scenarios. Though some initial efforts have been taken to interpret the predictions of GNNs, they mainly focus on providing post hoc explanations using an additional explainer, which could misrepresent the true inner working mechanism of the target GNN. The works on self-explainable GNNs are rather limited. Therefore, we study a novel problem of learning prototype-based self-explainable GNNs that can simultaneously give accurate predictions and prototype-based explanations on predictions. We design a framework which can learn prototype graphs that capture representative patterns of each class as class-level explanations. The learned prototypes are also used to simultaneously make prediction for a test instance and provide instance-level explanation. Extensive experiments on real-world and synthetic datasets show the effectiveness of the proposed framework for both prediction accuracy and explanation quality.
Graph Neural Networks (GNNs) have achieved great success in modeling graph-structured data. However, recent works show that GNNs are vulnerable to adversarial attacks which can fool the GNN model to make desired predictions of the attacker. In addition, training data of GNNs can be leaked under membership inference attacks. This largely hinders the adoption of GNNs in high-stake domains such as e-commerce, finance and bioinformatics. Though investigations have been made in conducting robust predictions and protecting membership privacy, they generally fail to simultaneously consider the robustness and membership privacy. Therefore, in this work, we study a novel problem of developing robust and membership privacy-preserving GNNs. Our analysis shows that Information Bottleneck (IB) can help filter out noisy information and regularize the predictions on labeled samples, which can benefit robustness and membership privacy. However, structural noises and lack of labels in node classification challenge the deployment of IB on graph-structured data. To mitigate these issues, we propose a novel graph information bottleneck framework that can alleviate structural noises with neighbor bottleneck. Pseudo labels are also incorporated in the optimization to minimize the gap between the predictions on the labeled set and unlabeled set for membership privacy. Extensive experiments on real-world datasets demonstrate that our method can give robust predictions and simultaneously preserve membership privacy.