Reachable sets of classifiers and regression models: (non-)robustness analysis and robust training

Keyphrases

• Regression Model 100%
• Robust Training 100%
• Reachable Set 100%
• Robustness Analysis 100%
• Classifier Model 100%
• Neural Network 66%
• Under-approximation 66%
• Reliability Prediction 66%
• Classification Task 33%
• State-of-the-art Techniques 33%
• Adversarial Attack 33%
• Explainability 33%
• Training Procedure 33%
• Over-under 33%
• Feature Ranking 33%
• Continuous Set 33%
• Norm Bounds 33%
• Robust Property 33%
• Adversarial States 33%
• Regression Task 33%

Computer Science

• Neural Network 100%
• Approximation (Algorithm) 100%
• Adversarial Machine Learning 50%
• Regression Task 50%
• Classification Task 50%