Open Theses

Requirements. All theses at ENCRYPTO are research-oriented. Therefore, we require that students who start a thesis at ENCRYPTO have successfully completed for a Bachelor thesis the course “Computersystemsicherheit” (or equivalent), and for a Master thesis at least one specialized ENCRYPTO course (CRYPROT, PRIVTECH, or PRIVDEV).

Below we give examples for open thesis topics, but often have further topics that are not announced and are also open to discuss innovative self-proposed topics that match with our research profile. To allow us to find a suitable topic, please send a list of courses you attended successfully, ideally in form of a current transcript of records.

For an idea what our theses topics can be about, check out currently running and finished theses.

  • Bachelor Thesis, Master Thesis

    The goal of this thesis is (1) to analyze the existing file formats for describing circuits for secure multi-party computation (MPC) and (2) to design a new well-defined, flexible file format and in-memory intermediate representation for describing MPC functionalities, circuits, and implementation-specific optimizations.

  • Master Thesis

    Privacy has become a crucial topic that nowadays is addressed in a wide range of domains outside of cryptography. It has been shown that such research proposing non-standard secure computation constructions can be seriously flawed, as fundamental impossibility results were not considered. This thesis aims at surveying the literature for such cases, identifying misuse of secure computation, and proposing specific attacks and, if possible, fixes.

  • Bachelor Thesis, Master Thesis

    This thesis should investigate and improve the efficiency of state-of-the-art privacy-preserving neural network training leveraging trusted execution environments (TEEs).

  • Bachelor Thesis, Master Thesis

    This thesis should investigate and improve the state-of-the-art in privacy-preserving clustering by designing a highly efficient and practical private clustering scheme that achieves good clustering quality.