📖 Course Overview
The course is designed to provide an introductory yet technical understanding of the security vulnerabilities and threats that modern AI systems face, with a focus on deep neural networks and computer vision applications. Participants will explore key definitions and techniques for designing and implementing AI-based security attacks and safety threats while also gaining a foundational understanding of how to protect AI systems against them. A portion of the course will be devoted to hands-on laboratory sessions, where participants will implement attacks and countermeasures in practical deep neural network applications.
The main topics of the course are organized as follows:
- AI Foundations, Threat Modeling, and preliminaries
- Adversarial Attacks, Defenses, and Robust Training
- Poisoning Attacks, Backdoors, and Dataset Analysis
- Privacy Attacks and Distributed Learning
- Explainable, Fairness, and Ethical AI
- Security and Safety of large models and Agentic Systems
🛠️ Format & Exam
- Lectures: 30 hours (in-person, TECIP Institute, CNR Area, Pisa).
- Exam (3 CFU): Project/Research work + oral discussion.
To attend the course in the first semester of the 2026–2027 academic year, please fill out the following form: Course Registration Form.
A Microsoft Teams channel will be set up soon for feedback, questions, and announcements.
🗓️ Schedule
TBD — Late October to November 2026 (First Semester)
📂 Lectures
📬 Contact
giulio.rossolini@santannapisa.it