Attacks and Defenses in Robust Machine Learning is an authoritative, deeply structured guide that explores the full spectrum of adversarial machine learning. Designed for engineers, researchers, cybersecurity experts, and policymakers, the book delivers critical insights into how modern AI systems can be compromised and how to protect them.
Spanning 30 chapters, it covers everything from adversarial theory and attack taxonomies to hands-on defense strategies across key domains like vision, NLP, healthcare, finance, and autonomous systems. With mathematical depth, real-world case studies, and forward-looking analysis, it balances rigor and practicality.
Ideal for:
- ML engineers and cybersecurity professionals building resilient systems
- Researchers and grad students studying adversarial ML
- Policy and tech leaders shaping AI safety and legal frameworks
Key features:
- In-depth coverage of attacks (evasion, poisoning, backdoors) and defenses (distillation, transformations, robust architectures)
- Sector-specific risks and mitigation strategies
- Exploration of privacy risks, legal implications, and future trends
This is the definitive resource for anyone aiming to understand and secure AI in an increasingly adversarial landscape.
