Artificial Intelligence

Artificial Intelligence: Data and Model Safety, 1st Edition, presents a structured technical foundation for securing modern computational systems against data manipulation and structural exploits. Written for graduate students studying computer science, artificial intelligence, or software engineering, the volume establishes essential principles for analyzing vulnerabilities across predictive platforms.

The book organizes its analytical coverage into complementary thematic sequences. Initial sections introduce fundamental machine learning mechanisms and basic safety concepts before detailing specific data safety attacks alongside corresponding defensive measures. The text then transitions directly to model security, evaluating how adversarial attacks compromise system stability and presenting computational techniques for adversarial defense.

Further chapters examine specialized threat categories, focusing on backdoor attacks, backdoor defenses, and model extraction risks. By systematically connecting attack mechanisms to defensive solutions, this publication provides graduate students with a structured framework for addressing technical security challenges in complex software environments.