Our research group conducts fundamental research at the intersection of computer security and machine learning. On the one end, we are interested in developing intelligent systems that can learn to protect computers from attacks and identify security problems automatically. On the other end, we explore the security and privacy of machine learning by developing novel attacks and defenses.
We are part of the new Berlin Institute for the Foundations of Learning and Data (BIFOLD). Previously, we have been working at Technische Universität Braunschweig and the University of Göttingen.
This lab is a hands-on course that explores machine learning in computer security. Students design and develop intelligent systems for security problems such as attack detection, malware clustering, and vulnerability discovery. The developed systems are trained and evaluated on real-world data, providing insight into their strengths and weaknesses in practice. The lab is a continuation of the lecture "Machine Learning for Computer Security" and thus knowledge from that course is expected.
STEMO — Steganography with Language Models
This project explores how large language modules, such as ChatGPT, can be used for steganography. Students will form a red team (attackers) and a blue team (defenders). The red team will develop techniques to hide secret messages in generated texts, while the blue team will develop methods to detect these messages. The color of the teams will change after some time. The project is aimed at Master students. A good understanding of language models and strong programming skills are required.
See all teaching course.
On the Detection of Image-Scaling Attacks in Machine Learning.
Proc. of the 39th Annual Computer Security Applications Conference (ACSAC), 2023. (to appear)
PAVUDI: Patch-based Vulnerability Discovery using Machine Learning.
Proc. of the 39th Annual Computer Security Applications Conference (ACSAC), 2023. (to appear)
Broken Promises: Measuring Confounding Effects in Learning-based Vulnerability Discovery.
Proc. of the 16th ACM Workshop on Artificial Intelligence and Security (AISEC), 2023. (to appear)
See all publications.
We are generally looking for motivated and skilled PhD students and postdocs to join our group—even if we currently do not announce open positions. If you are passionate about research and interested in combining machine learning and computer security, contact us directly at jobs@mlsec.org.
Before writing an unsolicited email, take some time to write a good cover letter. In this letter, you should describe why you are a good fit for our group and what research you enjoy doing. Include the result of (0x62df**215)%0xf0e5 in the subject line of your email.
Technische Universität Berlin
Machine Learning and Security, TEL 8-2
Ernst-Reuter-Platz 7
10587 Berlin, Germany
Responsibility under the German Press Law §55 Sect. 2 RStV:
Prof. Dr. Konrad Rieck
Email: rieck@tu-berlin.de