The secret revealer: Generative model-inversion attacks against deep neural networks

Keyphrases

• Generative Models 100%
• Deep Neural Network 100%
• Model Inversion Attack 100%
• Training Data 75%
• Predictive Models 50%
• Neural Network 25%
• Strong Correlation 25%
• Inversion Method 25%
• Logistic Regression 25%
• Vulnerability 25%
• Linear Regression 25%
• Face Recognition 25%
• Predictive Power 25%
• Identification Accuracy 25%
• Face Image 25%
• Canonical Form 25%
• Adversary 25%
• Public Information 25%
• High Failure Rate 25%
• Differential Privacy 25%
• Attack Strategy 25%
• Avail 25%
• Privacy Data 25%
• Generative Adversarial Networks 25%
• Novel Attacks 25%
• Inversion Attack 25%
• Private Training 25%

Computer Science

• Training Data 100%
• Generative Model 100%
• Deep Neural Network 100%
• Neural Network 33%
• Face Recognition 33%
• Predictive Power 33%
• Predictive Model 33%
• Sensitive Informations 33%
• Logistic Regression 33%
• Differential Privacy 33%
• Canonical Form 33%
• Generative Adversarial Networks 33%