Transferring Robustness for Graph Neural Network Against Poisoning Attacks
Abstract
Graph neural networks (GNNs) are widely used in many applications. However, their robustness against adversarial attacks is criticized. Prior studies show that using unnoticeable modifications on graph topology or nodal features can significantly reduce the performances of GNNs. It is very challenging to design robust graph neural networks against poisoning attack and several efforts have been taken. Existing work aims at reducing the negative impact from adversarial edges only with the poisoned graph, which is sub-optimal since they fail to discriminate adversarial edges from normal ones. On the other hand, clean graphs from similar domains as the target poisoned graph are usually available in the real world. By perturbing these clean graphs, we create supervised knowledge to train the ability to detect adversarial edges so that the robustness of GNNs is elevated. However, such potential for clean graphs is neglected by existing work. To this end, we investigate a novel problem of improving the robustness of GNNs against poisoning attacks by exploring clean graphs. Specifically, we propose PA-GNN, which relies on a penalized aggregation mechanism that directly restrict the negative impact of adversarial edges by assigning them lower attention coefficients. To optimize PA-GNN for a poisoned graph, we design a meta-optimization algorithm that trains PA-GNN to penalize perturbations using clean graphs and their adversarial counterparts, and transfers such ability to improve the robustness of PA-GNN on the poisoned graph. Experimental results on four real-world datasets demonstrate the robustness of PA-GNN against poisoning attacks on graphs. Code and data are available here: https://github.com/tangxianfeng/PA-GNN.
- Publication:
-
arXiv e-prints
- Pub Date:
- August 2019
- DOI:
- 10.48550/arXiv.1908.07558
- arXiv:
- arXiv:1908.07558
- Bibcode:
- 2019arXiv190807558T
- Keywords:
-
- Computer Science - Machine Learning;
- Computer Science - Cryptography and Security;
- Computer Science - Social and Information Networks;
- Statistics - Machine Learning
- E-Print:
- Accepted by WSDM 2020. Code and data: https://github.com/tangxianfeng/PA-GNN