Summer Semester 2025

Content

In this seminar, different security aspects of artificial intelligence, systems, and embedded devices will be analyzed and discussed. Students will process, summarize, and evaluate a number of current scientific publications for a certain topic in the form of an essay. Additionally, each student will present his work in front of the group at the end of the semester.

Possible topics include:

• Security models of current smartphone operating systems (e.g. Android, iOS, Windows Phone, MeeGo, Symbian, RIM)
• Security of mobile devices
• Trusted Hardware
• Internet of Things (IoT)
• Application security (e.g., mobile malware and runtime attacks)
• Privacy aspects in mobile devices
• Security of mobile networks
• Applications of Machine Learning for Security
• Privacy aspects of Deep Neural Networks
• Security Attacks and Defenses against Deep Neural Networks
• Distributed Training of Deep Neural Networks

Please check Moodle for full information:

https://moodle.informatik.tu-darmstadt.de/course/view.php?id=1727

Content

This course covers current security topics in research and development.

Analyzing and developing security solutions are complex tasks requiring knowledge from different areas of computer science. This lab aims to combine competencies from different areas within the framework of a security project.

As part of this course, tasks from a broad spectrum (including algorithms, space travel, machine learning, Deep Neural Networks, software analysis, hardware development, and reverse engineering) will be presented.

The individual tasks will be determined individually and according to the interests/skills of the participants.

Depending on the scope and level of the task, this course will be completed as a Lab (InoSys Lab with 6CP) or as a Practical Project (InoSys Project with 9CP). This type is determined individually and task-specifically. When choosing between the two types, and if the nature of the task allows it, students have the opportunity to participate intellectually in the design of the task.

Please check Moodle for full information:

https://moodle.informatik.tu-darmstadt.de/course/view.php?id=1752

To be able to handle more and more complex tasks, Deep Neural Networks need to be trained with more and more data. One strategy to obtain a large dataset in times of increasing privacy awareness is Federated Learning. Instead of requiring all data on one central place, Federated Learning outsources the training to several clients, such that each client trains its own model locally and only shares the parameters of the trained models.

The lectures will be accompanied by practical homeworks, where state-of-the-art attacks and defenses will be implemented. At the end, you will implement your own attack and defenses that have to compete against the submissions of the other participants.

Content:

• Challenges of using Machine Learning for Security Critical Tasks
• Introduction to different strategies for collaborative learning, including Split Learning and Federated Learning
• Privacy Challenges in Federated Learning
• Security attacks in Federated Learning (also-called Poisoning Attacks)
• State-of-the-art Poisoning Attacks and Defenses in Federated Learning

Please check Moodle for full information:

https://moodle.informatik.tu-darmstadt.de/course/view.php?id=1747

Content

• Trusted Computing
  
  • - Authenticated Boot
  • - Binding and Sealing
  • - Integrity Measurement and Attestation
  • - Direct Anonymous Attestation
  • - On-board Credentials
• Mobile Security with focus on smartphones
  
  • - Security Architectures
  • - Selected Access Control and Permission Model Aspects
  • - Context-based Security Policies
  • - Selected Modern Attack Techniques
• Hardware-based Cryptography
  
  • - Hardware-assisted Cryptographic Protocols
  • - Introduction to Physical Unclonable Functions (PUFs)

Please check Moodle for full information:

https://moodle.informatik.tu-darmstadt.de/course/view.php?id=1746