MedAttacker: Exploring Black-Box Adversarial Attacks on Risk Prediction Models in Healthcare

Computer Science

• Prediction Model 100%
• Adversarial Machine Learning 100%
• Reinforcement Learning 50%
• Selection Process 25%
• Learning Framework 25%
• Deep Neural Network 25%
• Temporal Context 25%
• Attack Technique 25%
• Selection Policy 25%

Keyphrases

• Position Selection 50%
• Hierarchical Positions 25%
• Contribution Index 25%
• Saliency Ratio 25%
• Services on the Cloud 25%
• Gray-box Setting 25%