Faculty
Mario Fritz is faculty member at the CISPA Helmholtz Center for Information Security and professor at the Saarland University. Before, he was senior researcher and research group head at the Max Planck Institute for Informatics, and PostDoc at the International Computer Science Institute and UC Berkeley. He did his PhD at the TU Darmstadt. His current work is centered around Trustworthy Information Processing with a focus on the intersection of AI \& Machine Learning with Security \& Privacy. He served as Area Chair for major computer vision conferences (ECCV, ICCV), associate editor of IEEE TPAMI and is a member of the ACM Europe Technical Policy Committee Europe. He has co-authored over 100 publications, including more than 50 in top-tier journals (IJCV, TPAMI) and conferences (CVPR, ICCV, ECCV, NeurIPS, AAAI, ICLR, NDSS, USENIX Security, CCS, S\& P). He is also a leading scientist of the Helmholtz Medical Security, Privacy, and AI Research Center, where he is coordinating projects on trustworthy federated data-analytics and protecting genetic data with synthetic cohorts from deep generative models.
SP
2020 IEEE Symposium on Security and Privacy (S&P)IEEE Symposium on Security and Privacy (S&P)
ICLR
8th International Conference on Learning RepresentationsInternational Conference on Representation Learning (ICLR)
Neurocomputing
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
ICCV
Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)International Conference on Computer Vision (ICCV)
ICCV
Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)International Conference on Computer Vision (ICCV)
CVPR
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Trustworthy Machine Learning
Machine learning has made great advances over the past year and many techniques have found their ways into applications. This leads to an increasing demand of techniques that not only perform well - but are also "trustworthy".
Trustworthiness includes:
- Interpretability of the prediction
- Robustness against changes to the input, which occur naturally or with malicious intend
- Privacy preserving machine learning (e.g. when dealing with sensitive data such as in health applications)
- Fairness
- ...
As a proseminar’s primary purpose is to learn presentation skills, the seminar will feature two presentations from each student.
In the first half of the semester, we will have presentations of two topics each week. After each presentation, fellow students and lecturers will provide feedback on how to improve the presentation. This general feedback must then be taken into account for the second half of the semester, where again each student will present.
The first presentations will count towards 30% of the overall grade, the second presentation will count towards 70% of the overall grade. Attendance in the proseminar meetings is mandatory. At most one session can be skipped, after that you need to bring a doctor’s note to excuse your absence.
Machine Learning in Cyber Security
Recent advances in Machine Learning has lead to near (or beyond) human-level performance in many tasks - autonomous driving, voice assistance, playing a variety of games. In terms of privacy and security, this is a double-edged sword. ML techniques can be used to efficiently detect and prevent attacks (e.g., intrusion detection). However, their deployment to many real-world sensitive systems (e.g., self-driving cars, the cloud) also makes them susceptible to numerous attacks, such as introducing imperceptible perturbations in inputs and forcing ML systems behave in unintended ways.
The course explores in-depth both of these sides to Machine Learning and Cyber Security. The content addresses the following areas:
While we do a brief recap in the beginning, the course requires knowledge on Machine Learning.
Date for lecture: Tuesdays noon to 2pm.
Date for exercise: Fridays 2pm to 4pm
Due to the size of the course - the lecture will start in an online format until further notice.
The course requires prior knowledge on Machine Learning.
Once you have registered - please find internal information and schedule and links here (under construction).
Proseminar: Trustworthy Machine Learning
Machine learning has made great advances over the past year and many techniques have found their ways into applications. This leads to an increasing demand of techniques that not only perform well - but are also "trustworthy".
Lecture: High Level Computer Vision
This course will cover essential techniques for high-level computer vision. These techniques facilitate semantic interpretation of visual data, as it is required for a broad range of applications like robotics, driver assistance, multi-media retrieval, surveillance etc.
Lecture: Machine Learning in Cybersecurity
Recent advances in Machine Learning has lead to near (or beyond) human-level performance in many tasks - autonomous driving, voice assistance, playing a variety of games.