Fast Gradient Attack on Network Embedding

TL;DR

This paper introduces a fast gradient attack method on network embedding models that effectively disrupts their performance by rewiring minimal links, raising privacy concerns in social network analysis.

Contribution

The paper proposes a novel Fast Gradient Attack (FGA) framework that leverages gradient information in GCNs to efficiently generate adversarial networks for attacking various embedding methods.

Findings

FGA outperforms baseline attack methods in disrupting network embeddings.

Only a few link modifications are needed to significantly impair embedding quality.

The method is effective across multiple real-world networks and embedding techniques.

Abstract

Network embedding maps a network into a low-dimensional Euclidean space, and thus facilitate many network analysis tasks, such as node classification, link prediction and community detection etc, by utilizing machine learning methods. In social networks, we may pay special attention to user privacy, and would like to prevent some target nodes from being identified by such network analysis methods in certain cases. Inspired by successful adversarial attack on deep learning models, we propose a framework to generate adversarial networks based on the gradient information in Graph Convolutional Network (GCN). In particular, we extract the gradient of pairwise nodes based on the adversarial network, and select the pair of nodes with maximum absolute gradient to realize the Fast Gradient Attack (FGA) and update the adversarial network. This process is implemented iteratively and terminated until certain condition is satisfied, i.e., the number of modified links reaches certain predefined value. Comprehensive attacks, including unlimited attack, direct attack and indirect attack, are performed on six well-known network embedding methods. The experiments on real-world networks suggest that our proposed FGA behaves better than some baseline methods, i.e., the network embedding can be easily disturbed using FGA by only rewiring few links, achieving state-of-the-art attack performance.