Adversarial Attack on Graph Neural Networks as An Influence Maximization Problem

Deng, Junwei; Mei, Qiaozhu

doi:10.48550/arxiv.2106.10785

Cited by 2 publications

(3 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Zügner et al [69,71] started the race of graph adversarial attacks. Pioneering works mainly focused on modifying node features [33,69,71] and perturbing edges [54,59,66,70]. Some recent works [5,7,20,21,48,51,55,68] study the node injection attack, which injects new nodes into a graph and connects them with some existing nodes.…”

Section: Related Work 21 Graph Adversarial Attackmentioning

confidence: 99%

“…Evaluating the change of GNN predictions caused by the change of input has been studied in graph adversarial attack and graph counterfactual explanation. Graph adversarial attack aims to maximally undermine GNN performance or change GNN prediction by perturbations to the input graph, which mainly include modifying node features [33,69,71], injecting nodes [5,7,20,48,51,55,68], or edge perturbation [54,59,66,70]. However, to the best of our knowledge, none of the adversarial attack methods utilizes node removal, since it is not common in the attack scenario.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fast Inference of Removal-Based Node Influence

Li,

Xiao,

Luo

et al. 2024

Proceedings of the ACM Web Conference 2024

View full text Add to dashboard Cite

Graph neural networks (GNNs) are widely utilized to capture the information spreading patterns in graphs. While remarkable performance has been achieved, there is a new trending topic of evaluating node influence. We propose a new method of evaluating node influence, which measures the prediction change of a trained GNN model caused by removing a node. A real-world application is, "In the task of predicting Twitter accounts' polarity, had a particular account been removed, how would others' polarity change?". We use the GNN as a surrogate model whose prediction could simulate the change of nodes or edges caused by node removal. Our target is to obtain the influence score for every node, and a straightforward way is to alternately remove every node and apply the trained GNN on the modified graph to generate new predictions. It is reliable but time-consuming, so we need an efficient method. The related lines of work, such as graph adversarial attack and counterfactual explanation, cannot directly satisfy our needs, since their problem settings are different. We propose an efficient, intuitive, and effective method, NOde-Removal-based fAst GNN inference (NORA), which uses the gradient information to approximate the node-removal influence. It only costs one forward propagation and one backpropagation to approximate the influence score for all nodes. Extensive experiments on six datasets and six GNN models verify the effectiveness of NORA. Our code is available at https://github.com/weikai-li/NORA.git. CCS CONCEPTS• Information systems → Web mining; • Computing methodologies → Neural networks.

show abstract

Section: Related Work 21 Graph Adversarial Attackmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Fast Inference of Removal-Based Node Influence

Li,

Xiao,

Luo

et al. 2024

Proceedings of the ACM Web Conference 2024

View full text Add to dashboard Cite

show abstract

“…Node Injection Poisoning Attacks (NIPA) uses a hierarchical reinforcement learning approach to sequentially manipulate the labels and links of the injected nodes. Recently, InfMax [20] formulates the adversarial attack on GNNs as an influence maximization problem.…”

Section: Adversarial Attack On Graphsmentioning

confidence: 99%

Adversarial Graph Contrastive Learning with Information Regularization

Feng¹,

Jing²,

Zhu³

et al. 2022

Preprint

View full text Add to dashboard Cite

Contrastive learning is an effective unsupervised method in graph representation learning. Recently, the data augmentation based contrastive learning method has been extended from images to graphs. However, most prior works are directly adapted from the models designed for images. Unlike the data augmentation on images, the data augmentation on graphs is far less intuitive and much harder to provide high-quality contrastive samples, which are the key to the performance of contrastive learning models. This leaves much space for improvement over the existing graph contrastive learning frameworks. In this work, by introducing an adversarial graph view and an information regularizer, we propose a simple but effective method, Adversarial Graph Contrastive Learning (ArieL), to extract informative contrastive samples within a reasonable constraint. It consistently outperforms the current graph contrastive learning methods in the node classification task over various real-world datasets and further improves the robustness of graph contrastive learning.

show abstract

Adversarial Attack on Graph Neural Networks as An Influence Maximization Problem

Cited by 2 publications

References 20 publications

Fast Inference of Removal-Based Node Influence

Fast Inference of Removal-Based Node Influence

Adversarial Graph Contrastive Learning with Information Regularization

Contact Info

Product

Resources

About