Practical AI Security - Harriet Farlow

As AI technologies become increasingly popular, bad actors have begun targeting them for disruption. This book introduces you to the burgeoning fields of machine learning hacking and AI security. You'll follow simple examples written in the Python programming language to learn about training and deploying a machine learning model, then dive into the cyberattacks that can force those models to malfunction, whether by disclosing sensitive information or incorrectly classifying a result, sometimes with devastating outcomes. Once you gain experience performing these attacks yourself, you'll explore the ecosystem of tools that exist to defend against them before touring the ways the industry tries to secure AI by assessing risk, implementing standards, and influencing legislation.
více informací