A simple unified framework for detecting out-of-distribution samples and adversarial attacks

Computer Science

• Application 33%
• Art Performance 33%
• Attack 33%
• Classes 100%
• Classification (Machine Learning) 66%
• Classification Rule 33%
• Confidence Score 33%
• Deep Neural Network 33%
• Discriminant Analysis 33%
• Effective Method 33%
• Good Classifier 33%
• Machine Learning 33%
• Mahalanobis Distance 33%
• Posterior Distribution 33%
• Real World 33%
• Sample Distribution 66%
• Simulation Mode 66%
• Training Dataset 33%
• Unified Framework 33%