Adversarial Robustness for Machine Learning summarizes the recent progress on this topic and introduces popular algorithms on adversarial attack, defense and veri?cation. Sections cover adversarial attack, veri?cation and defense, mainly focusing on image classi?cation applications which are the standard benchmark considered in the adversarial robustness community. Other sections discuss adversarial examples beyond image classification, other threat models beyond testing time attack, and applications on adversarial robustness. For researchers, this book provides a thorough literature review that summarizes latest progress in the area, which can be a good reference for conducting future research.
In addition, the book can also be used as a textbook for graduate courses on adversarial robustness or trustworthy machine learning. While machine learning (ML) algorithms have achieved remarkable performance in many applications, recent studies have demonstrated their lack of robustness against adversarial disturbance. The lack of robustness brings security concerns in ML models for real applications such as self-driving cars, robotics controls and healthcare systems.
1. Adversarial Robustness for Machine Learning - p. iii 2. Cover - p. iii 3. Title Page - p. iii 4. Copyright Page - p. iv 5. Table of Contents - p. x 6. Dedication - p. v 7. Biography - p. xi 8. Preface - p. xiii 9. Chapter 1 Background and motivation - p. 3 10. Chapter 2 White-box adversarial attacks - p. 15 11. Chapter 3 Black-box adversarial attacks - p. 29 12. Chapter 4 Physical adversarial attacks - p. 47 13. Chapter 5 Training-time adversarial attacks - p. 51 14. Chapter 6 Adversarial attacks beyond image classification - p. 59 15. Chapter 7 Overview of neural network verification - p. 73 16. Chapter 8 Incomplete neural network verification - p. 79 17. Chapter 9 Complete neural network verification - p. 89 18. Chapter 10 Verification against semantic perturbations - p. 95 19. Chapter 11 Overview of adversarial defense - p. 113 20. Chapter 12 Adversarial training - p. 119 21. Chapter 13 Randomization-based defense - p. 127 22. Chapter 14 Certified robustness training - p. 137 23. Chapter 15 Adversary detection - p. 143 24. Chapter 16 Adversarial robustness of beyond neural network models - p. 157 25. Chapter 17 Adversarial robustness in meta-learning and contrastive learning - p. 183 26. Chapter 18 Model reprogramming - p. 201 27. Chapter 19 Contrastive explanations - p. 217 28. Chapter 20 Model watermarking and fingerprinting - p. 227 29. Chapter 21 Data augmentation for unsupervised machine learning - p. 241 30. References - p. 251 31. Index - p. 273
Adversarial Robustness for Machine Learning summarizes the recent progress on this topic and introduces popular algorithms on adversarial attack, defense and veri?cation. Sections cover adversarial attack, veri?cation and defense, mainly focusing on image classi?cation applications which are the standard benchmark considered in the adversarial robustness community. Other sections discuss adversarial examples beyond image classification, other threat models beyond testing time attack, and applications on adversarial robustness. For researchers, this book provides a thorough literature review that summarizes latest progress in the area, which can be a good reference for conducting future research.
In addition, the book can also be used as a textbook for graduate courses on adversarial robustness or trustworthy machine learning. While machine learning (ML) algorithms have achieved remarkable performance in many applications, recent studies have demonstrated their lack of robustness against adversarial disturbance. The lack of robustness brings security concerns in ML models for real applications such as self-driving cars, robotics controls and healthcare systems.
1. Adversarial Robustness for Machine Learning - p. iii 2. Cover - p. iii 3. Title Page - p. iii 4. Copyright Page - p. iv 5. Table of Contents - p. x 6. Dedication - p. v 7. Biography - p. xi 8. Preface - p. xiii 9. Chapter 1 Background and motivation - p. 3 10. Chapter 2 White-box adversarial attacks - p. 15 11. Chapter 3 Black-box adversarial attacks - p. 29 12. Chapter 4 Physical adversarial attacks - p. 47 13. Chapter 5 Training-time adversarial attacks - p. 51 14. Chapter 6 Adversarial attacks beyond image classification - p. 59 15. Chapter 7 Overview of neural network verification - p. 73 16. Chapter 8 Incomplete neural network verification - p. 79 17. Chapter 9 Complete neural network verification - p. 89 18. Chapter 10 Verification against semantic perturbations - p. 95 19. Chapter 11 Overview of adversarial defense - p. 113 20. Chapter 12 Adversarial training - p. 119 21. Chapter 13 Randomization-based defense - p. 127 22. Chapter 14 Certified robustness training - p. 137 23. Chapter 15 Adversary detection - p. 143 24. Chapter 16 Adversarial robustness of beyond neural network models - p. 157 25. Chapter 17 Adversarial robustness in meta-learning and contrastive learning - p. 183 26. Chapter 18 Model reprogramming - p. 201 27. Chapter 19 Contrastive explanations - p. 217 28. Chapter 20 Model watermarking and fingerprinting - p. 227 29. Chapter 21 Data augmentation for unsupervised machine learning - p. 241 30. References - p. 251 31. Index - p. 273
