Towards an efficient and general framework of robust training for graph neural networks

Keyphrases

• Adversarial Examples 20%
• Computationally Expensive 20%
• Convex Relaxation 20%
• Design Training 20%
• Discrete Graph 20%
• Gradient Method 20%
• Graph Datasets 20%
• Graph Domain 20%
• Graph Neural Network 100%
• Graph Structure 20%
• Greedy Search Algorithm 20%
• Inference Tasks 20%
• Large Graphs 20%
• Large-scale Problems 20%
• Neural Network Design 20%
• Node Attributes 20%
• Regret 20%
• Robust Graph 40%
• Robust Training 100%
• Security Concerns 20%
• State-of-the-art Techniques 20%
• Structure Variable 20%
• Task Graph 20%
• Time-sensitive Tasks 20%
• Training Methods 40%
• White-box 20%
• Zeroth-order Methods 20%

Computer Science

• Adversarial Example 20%
• Convex Relaxation 20%
• Graph Neural Network 100%
• Inference Task 20%
• Large-Scale Problem 20%
• Network Design 20%
• Searching Algorithm 20%
• Security Concern 20%
• Time-Sensitive 20%