Unsupervised graph poisoning attack via contrastive loss back-propagation

Keyphrases

• Backpropagation 100%
• Graph Contrastive Learning 100%
• Contrastive Loss 100%
• Poisoning Attack 100%
• Downstream Task 33%
• Adversarial Attack 33%
• Copyright 16%
• Learning Model 16%
• Computing Machinery 16%
• Adjacency Matrix 16%
• Learning Framework 16%
• Graph Representation 16%
• Two-view 16%
• Supervised Approach 16%
• Gradient-based 16%
• Specific Scenario 16%
• Representation Method 16%
• Classification Prediction 16%
• Unsupervised Graph Representation Learning 16%
• Gradient Ascent 16%
• Graph Adversarial Attack 16%
• Supervised Model 16%
• Node-Link 16%
• Node Classification 16%
• Scenario Making 16%
• Link Prediction 16%
• Graph Representation Model 16%
• Attack Strategy 16%

Computer Science

• Backpropagation 100%
• Contrastive Learning 100%
• Contrastive Loss 100%
• Adversarial Machine Learning 50%
• Graph Representation 33%
• Computing Machinery 16%
• Adjacency Matrix 16%
• Learning Framework 16%
• Representation Model 16%
• Representation Method 16%
• Graph Representation Learning 16%
• Link Prediction 16%
• Node Classification 16%