After successful completion of the course, students are able to...
After completing this course, students will be equipped with a thorough understanding of how artificial intelligence (AI) can be applied to tackle security problems. They can employ learning algorithms to address specific security
tasks, including attack detection and malware analysis. Furthermore, the course will provide them with insights into the roles of data selection, feature extraction, and learning techniques and how these elements influence the
success of AI in security applications. Additionally, students will learn about the limitations, challenges, and potential oversights inherent in deploying AI for security purposes.
The lecturer of this course will be Konrad Rieck / TU Berlin.
This course provides an introduction to artificial intelligence in security applications. Rather than jumping on the hype train of AI, we first look back at previous research and discuss general concepts for applying learning
algorithms to security tasks, such as attack detection and malware analysis. We then examine the role of data, features, and learning techniques in this context and address their limitations and blind spots. The course concludes with an
outlook on future developments and possible misconceptions in AI-based security.
- Basics of Machine Learning (4h)
- Features and Feature Spaces (2h)
- Attack Detection using Machine Learning (4h)
- Malware Analysis using Machine Learning (2h)
- Robustness and Adversarial Learning (2h)
- Generative AI in Security Applications (2h)
- Critical Reflections and Outlook (4h)
Planned schedule:
Block course with two separate weeks in the semester:
Week 1: 13.5 - 17.5., 5x2 hours
Week 2: 24.6. - 28.6., 5x2 hours
The course format is a lecture combined with interactive discussions. Each unit will focus on a single topic, structured to include 2/3 of the time dedicated to instructive teaching and 1/3 allocated for engaging in discussions with the students.
