来源：深度学习与图网络本文约1400字，建议阅读5分钟本文为你分享图上的对抗与攻击精选论文。

2021相关论文一览

1. 大规模攻击图神经网络

2. 图神经网络的黑盒梯度攻击：

   更深入洞察图的攻击和防御

3. 增强多路复用网络对节点社区级联故障的鲁棒性和弹性

4. PATHATTACK：

   攻击复杂网络中的最短路径

5. Deformable shape的通用谱对抗攻击

6. Preserve, Promote, or Attack?通过拓扑扰动的 GNN 解释

7. 网络嵌入攻击：

   一种基于欧几里德距离的方法

8. 通过监督网络Poisoning对网络嵌入的对抗性攻击

9. DeHiB：

   通过对抗性扰动对半监督学习的深层隐藏后门攻击

10. GraphAttacker：

    一个通用的多任务图攻击框架

11. 图神经网络的成员推理攻击

1. Attacking Graph Neural Networks at Scale

2. Black-box Gradient Attack on Graph Neural Networks: Deeper Insights in Graph-based Attack and Defense

3. Enhancing Robustness and Resilience of Multiplex Networks Against Node-Community Cascading Failures

4. PATHATTACK: Attacking Shortest Paths in Complex Networks

5. Universal Spectral Adversarial Attacks for Deformable Shapes

6. Preserve, Promote, or Attack? GNN Explanation via Topology Perturbation

7. Network Embedding Attack: An Euclidean Distance Based Method

8. Adversarial Attack on Network Embeddings via Supervised Network Poisoning

9. DeHiB: Deep Hidden Backdoor Attack on Semi-Supervised Learning via Adversarial Perturbation

10. GraphAttacker: A General Multi-Task Graph Attack Framework

11. Membership Inference Attack on Graph Neural Networks

2020年相关研究论文

1. 图神经网络的对抗性标签翻转攻击和防御

2. 对图神经网络的探索性对抗攻击

3. 对图卷积网络的有针对性的通用攻击

4. 在不改变现有连接的情况下攻击基于图的分类

5. 学习通过有针对性的扰动欺骗知识图谱增强模型

6. 基于图神经网络的时空预测的一种顶点攻击

7. 欺骗图神经网络的单节点攻击

8. 图神经网络的黑盒对抗攻击作为影响最大化问题

9. 深度图匹配的对抗性攻击

10. 对图神经网络进行Practical对抗性攻击

11. 一种对隐私保护记录链接的图匹配攻击

12. 通过 GAN 对图嵌入的自适应对抗性攻击

13. 乘法器交替方向法对图神经网络的可扩展对抗性攻击

14. 针对用于恶意软件检测的图神经网络的语义保留强化学习攻击

15. 对大规模图的对抗性攻击

16. 通过影响函数(Influence Function)对图神经网络进行有效的规避攻击

17. 基于强化学习的黑盒规避攻击在动态图中进行链接预测

18. 针对无标度网络的 BC 分类的对抗性攻击

19. 基于图神经网络的链路预测算法的对抗性攻击

20. 图神经网络的Practical对抗性攻击

21. 通过迭代梯度攻击的链路预测对抗性攻击

22. 对图结构化数据的有效对抗性攻击

23. 图Backdoor

24. 图神经网络的Backdoor攻击

25. 通过 Nash 强化学习进行垃圾邮件发送检测

26. 图神经网络的对抗性攻击：扰动及其模式

27. 对分层图池化神经网络的对抗性攻击

28. 从图神经网络窃取链接

29. 通过注入恶意节点对图数据进行可扩展攻击

30. 网络中断：最大化社交网络中的分歧和两极分化

31. 网络中意见动态的对抗性扰动

32. 图神经网络上的非目标特定节点注入攻击：一种分层强化学习方法

33. MGA：网络上的动量梯度攻击

34. 通过对图卷积网络进行Poisoning邻居的间接对抗性攻击

35. 图通用对抗性攻击：一些不良行为者破坏图学习模型

36. 对无标度网络的对抗性攻击：测试物理标准的稳健性

37. 通过隐藏个人对社区检测的对抗性攻击

1. Adversarial Label-Flipping Attack and Defense for Graph Neural Networks

2. Exploratory Adversarial Attacks on Graph Neural Networks

3. A Targeted Universal Attack on Graph Convolutional Network

4. Attacking Graph-Based Classification without Changing Existing Connections

5. Learning to Deceive Knowledge Graph Augmented Models via Targeted Perturbation

6. One Vertex Attack on Graph Neural Networks-based Spatiotemporal Forecasting

7. Single-Node Attack for Fooling Graph Neural Networks

8. Black-Box Adversarial Attacks on Graph Neural Networks as An Influence Maximization Problem

9. Adversarial Attacks on Deep Graph Matching  | Attack  |  Graph Matching  | Deep Graph Matching Models

10. Towards More Practical Adversarial Attacks on Graph Neural Networks

11. A Graph Matching Attack on Privacy-Preserving Record Linkage

12. Adaptive Adversarial Attack on Graph Embedding via GAN

13. Scalable Adversarial Attack on Graph Neural Networks with Alternating Direction Method of Multipliers

14. Semantic-preserving Reinforcement Learning Attack Against Graph Neural Networks for Malware Detection

15. Adversarial Attack on Large Scale Graph

16. Efficient Evasion Attacks to Graph Neural Networks via Influence Function

17. Reinforcement Learning-based Black-Box Evasion Attacks to Link Prediction in Dynamic Graphs

18. Adversarial attack on BC classification for scale-free networks

19. Adversarial Attacks on Link Prediction Algorithms Based on Graph Neural Networks

20. Practical Adversarial Attacks on Graph Neural Networks

21. Link Prediction Adversarial Attack Via Iterative Gradient Attack

22. An Efficient Adversarial Attack on Graph Structured Data

23. Graph Backdoor  | Attack  |  Node Classification Graph Classification

24. Backdoor Attacks to Graph Neural Networks

25. Robust Spammer Detection by Nash Reinforcement Learning

26. Adversarial Attacks on Graph Neural Networks: Perturbations and their Patterns

27. Adversarial Attack on Hierarchical Graph Pooling Neural Networks

28. Stealing Links from Graph Neural Networks

29. Scalable Attack on Graph Data by Injecting Vicious Nodes

30. Network disruption: maximizing disagreement and polarization in social networks

31. Adversarial Perturbations of Opinion Dynamics in Networks

32. Non-target-specific Node Injection Attacks on Graph Neural Networks: A Hierarchical Reinforcement Learning Approach

33. MGA: Momentum Gradient Attack on Network  | Attack  |  Node Classification, Community Detection

34. Indirect Adversarial Attacks via Poisoning Neighbors for Graph Convolutional Networks

35. Graph Universal Adversarial Attacks: A Few Bad Actors Ruin Graph Learning Models

36. Adversarial Attacks to Scale-Free Networks: Testing the Robustness of Physical Criteria

37. Adversarial Attack on Community Detection by Hiding Individuals

更多论文请查看:

https://github.com/safe-graph/graph-adversarial-learning-literature

编辑：文婧