In our BMBF-funded project Privacy E2E we are working on privacy-preserving AI systems and techniques to increase the privacy of citizens in the processing of their data without losing the added value of AI systems.
In the context of this project, the objective of the thesis is to develop a monitoring framework that allows the embedding of privacy monitors into AI-enabled systems. The monitoring framework should allow automated monitoring of specific AI systems based on specifications of what is privacy-critical data in the system. The monitoring must be based on a comparison with the previous states when retraining the ML models and continuously monitoring at runtime.
Prerequisites: Knowledge of machine learning and programming, preferably knowledge about privacy
Contact: Yorick Sens (yorick.sens@rub.de), Sven Peldszus (sven.peldszus@rub.de)
Extent: M.Sc.