Prerequisites: MET CS 767 or knowledge of Neural Networks or instructor’s consent. - This course is designed to provide students with a comprehensive understanding of the inherent vulnerabilities/security issues associated with integrating machine learning into various applications, as well as the knowledge/skills to defeat those vulnerabilities. Topics include an overview of categories of attacks against machine learning models and a detailed exploration of adversarial attacks, data poisoning attacks, membership inference attacks, model stealing attacks, as well as various defense solutions against the above-mentioned attacks. Upon the completion of the course, students are expected to know the threats and vulnerabilities that machine learning models face, along with the strategies and tools used to mitigate those risks. Hands-on labs based on existing tools are provided and required. 

Note that this information may change at any time. Please visit the MyBU Student Portal for the most up-to-date course information.