MGA: Momentum Gradient Attack on Network

Abstract

The adversarial attack methods based on gradient information can adequately find the perturbations, that is, the combinations of rewired links, thereby reducing the effectiveness of the deep learning model based graph embedding algorithms, but it is also easy to fall into a local optimum. Therefore, this paper proposes a Momentum Gradient Attack (MGA) against the GCN model, which can achieve more aggressive attacks with fewer rewiring links. Compared with directly updating the original network using gradient information, integrating the momentum term into the iterative process can stabilize the updating direction, which makes the model jump out of poor local optimum and enhance the method with stronger transferability. Experiments on node classification and community detection methods based on three well-known network embedding algorithms show that MGA has a better attack effect and transferability.

References in corpus (19)

• Fast unfolding of communities in large networks
• Finding and evaluating community structure in networks
• Modularity and community structure in networks
• Semi-Supervised Classification with Graph Convolutional Networks
• Fast algorithm for detecting community structure in networks
• An overview of gradient descent optimization algorithms
• Near linear time algorithm to detect community structures in large-scale networks
• Adversarial Attacks on Neural Networks for Graph Data
• Discriminative Embeddings of Latent Variable Models for Structured Data
• The Space of Transferable Adversarial Examples
• Community detection in networks: Modularity optimization and maximum likelihood are equivalent
• Poisoning Attacks to Graph-Based Recommender Systems
• GraphGAN: Graph Representation Learning with Generative Adversarial Nets
• Adversarial Attack on Graph Structured Data
• Fast Gradient Attack on Network Embedding
• Energy-entropy competition and the effectiveness of stochastic gradient descent in machine learning
• Gradient-based Optimization for Regression in the Functional Tensor-Train Format
• Curriculum Adversarial Training
• Target Defense Against Link-Prediction-Based Attacks via Evolutionary Perturbations