IDEA: Invariant Defense for Graph Adversarial Robustness
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917649390239744 |
|---|---|
| author | Tao, Shuchang Cao, Qi Shen, Huawei Wu, Yunfan Xu, Bingbing Cheng, Xueqi |
| author_facet | Tao, Shuchang Cao, Qi Shen, Huawei Wu, Yunfan Xu, Bingbing Cheng, Xueqi |
| contents | Despite the success of graph neural networks (GNNs), their vulnerability to adversarial attacks poses tremendous challenges for practical applications. Existing defense methods suffer from severe performance decline under unseen attacks, due to either limited observed adversarial examples or pre-defined heuristics. To address these limitations, we analyze the causalities in graph adversarial attacks and conclude that causal features are key to achieve graph adversarial robustness, owing to their determinedness for labels and invariance across attacks. To learn these causal features, we innovatively propose an Invariant causal DEfense method against adversarial Attacks (IDEA). We derive node-based and structure-based invariance objectives from an information-theoretic perspective. IDEA ensures strong predictability for labels and invariant predictability across attacks, which is provably a causally invariant defense across various attacks. Extensive experiments demonstrate that IDEA attains state-of-the-art defense performance under all five attacks on all five datasets. The implementation of IDEA is available at https://anonymous.4open.science/r/IDEA. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2305_15792 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | IDEA: Invariant Defense for Graph Adversarial Robustness Tao, Shuchang Cao, Qi Shen, Huawei Wu, Yunfan Xu, Bingbing Cheng, Xueqi Machine Learning Cryptography and Security Despite the success of graph neural networks (GNNs), their vulnerability to adversarial attacks poses tremendous challenges for practical applications. Existing defense methods suffer from severe performance decline under unseen attacks, due to either limited observed adversarial examples or pre-defined heuristics. To address these limitations, we analyze the causalities in graph adversarial attacks and conclude that causal features are key to achieve graph adversarial robustness, owing to their determinedness for labels and invariance across attacks. To learn these causal features, we innovatively propose an Invariant causal DEfense method against adversarial Attacks (IDEA). We derive node-based and structure-based invariance objectives from an information-theoretic perspective. IDEA ensures strong predictability for labels and invariant predictability across attacks, which is provably a causally invariant defense across various attacks. Extensive experiments demonstrate that IDEA attains state-of-the-art defense performance under all five attacks on all five datasets. The implementation of IDEA is available at https://anonymous.4open.science/r/IDEA. |
| title | IDEA: Invariant Defense for Graph Adversarial Robustness |
| topic | Machine Learning Cryptography and Security |
| url | https://arxiv.org/abs/2305.15792 |