Robust Mid-Pass Filtering Graph Convolutional Networks
Huang, Jincheng, Du, Lun, Chen, Xu, Fu, Qiang, Han, Shi, Zhang, Dongmei
–arXiv.org Artificial Intelligence
Graph convolutional networks (GCNs) are currently the most promising paradigm for dealing with graph-structure data, while recent studies have also shown that GCNs are vulnerable to adversarial attacks. Thus developing GCN models that are robust to such attacks become a hot research topic. However, the structural purification learning-based or robustness constraints-based defense GCN methods are usually designed for specific data or attacks, and introduce additional objective that is not for classification. Extra training overhead is also required in their design. To address these challenges, we conduct in-depth explorations on mid-frequency signals on graphs and propose a simple yet effective Mid-pass filter GCN (Mid-GCN). Theoretical analyses guarantee the robustness of signals through the mid-pass filter, and we also shed light on the properties of different frequency signals under adversarial attacks. Extensive experiments on six benchmark graph data further verify the effectiveness of our designed Mid-GCN in node classification accuracy compared to state-of-the-art GCNs under various adversarial attack strategies.
arXiv.org Artificial Intelligence
Feb-15-2023
- Country:
- North America
- United States
- Texas > Travis County
- Austin (0.05)
- New York > New York County
- New York City (0.04)
- Louisiana > Orleans Parish
- New Orleans (0.04)
- Texas > Travis County
- Canada > British Columbia
- United States
- Europe
- Asia > China
- Beijing > Beijing (0.05)
- Sichuan Province > Chengdu (0.04)
- Jiangsu Province > Yancheng (0.04)
- Africa > Ethiopia
- Addis Ababa > Addis Ababa (0.04)
- North America
- Genre:
- Research Report (1.00)
- Industry:
- Information Technology > Security & Privacy (1.00)
- Government > Military (0.97)
- Technology: