Adversarial attacks and defenses in machine learning-empowered communication systems and networks: A contemporary survey
Adversarial attacks and defenses in machine learning and deep neural network (DNN) have
been gaining significant attention due to the rapidly growing applications of deep learning in …
been gaining significant attention due to the rapidly growing applications of deep learning in …
Improving the transferability of adversarial samples by path-augmented method
Deep neural networks have achieved unprecedented success on diverse vision tasks.
However, they are vulnerable to adversarial noise that is imperceptible to humans. This …
However, they are vulnerable to adversarial noise that is imperceptible to humans. This …
Knowledge distillation improves graph structure augmentation for graph neural networks
Graph (structure) augmentation aims to perturb the graph structure through heuristic or
probabilistic rules, enabling the nodes to capture richer contextual information and thus …
probabilistic rules, enabling the nodes to capture richer contextual information and thus …
Extracting low-/high-frequency knowledge from graph neural networks and injecting it into mlps: An effective gnn-to-mlp distillation framework
Recent years have witnessed the great success of Graph Neural Networks (GNNs) in
handling graph-related tasks. However, MLPs remain the primary workhorse for practical …
handling graph-related tasks. However, MLPs remain the primary workhorse for practical …
Towards reasonable budget allocation in untargeted graph structure attacks via gradient debias
It has become cognitive inertia to employ cross-entropy loss function in classification related
tasks. In the untargeted attacks on graph structure, the gradients derived from the attack …
tasks. In the untargeted attacks on graph structure, the gradients derived from the attack …
Feature‐Based Graph Backdoor Attack in the Node Classification Task
Y Chen, Z Ye, H Zhao, Y Wang - International Journal of …, 2023 - Wiley Online Library
Graph neural networks (GNNs) have shown significant performance in various practical
applications due to their strong learning capabilities. Backdoor attacks are a type of attack …
applications due to their strong learning capabilities. Backdoor attacks are a type of attack …
Imperceptible graph injection attack on graph neural networks
Y Chen, Z Ye, Z Wang, H Zhao - Complex & Intelligent Systems, 2024 - Springer
Abstract In recent years, Graph Neural Networks (GNNs) have achieved excellent
applications in classification or prediction tasks. Recent studies have demonstrated that …
applications in classification or prediction tasks. Recent studies have demonstrated that …
Safety in Graph Machine Learning: Threats and Safeguards
Graph Machine Learning (Graph ML) has witnessed substantial advancements in recent
years. With their remarkable ability to process graph-structured data, Graph ML techniques …
years. With their remarkable ability to process graph-structured data, Graph ML techniques …
Learning to augment graph structure for both homophily and heterophily graphs
Recent years have witnessed great successes in performing graph structure learning for
Graph Neural Networks (GNNs). However, comparatively little work studies structure …
Graph Neural Networks (GNNs). However, comparatively little work studies structure …
A Black-box Adversarial Attack Method via Nesterov Accelerated Gradient and Rewiring Towards Attacking Graph Neural Networks
S Zhao, W Wang, Z Du, J Chen… - IEEE Transactions on Big …, 2023 - ieeexplore.ieee.org
Recent studies have shown that Graph Neural Networks (GNNs) are vulnerable to well-
designed and imperceptible adversarial attack. Attacks utilizing gradient information are …
designed and imperceptible adversarial attack. Attacks utilizing gradient information are …