Scalable Cryptographic Techniques for Distributed and Differentially Private Data Analysis

My research interests lie at the intersection of cryptography and data privacy. Combining these techniques is powerful and especially useful in distributed settings like machine learning, where models are trained on sensitive user data by companies that are not necessarily trustworthy. In my research, I design algorithms and protocols to enable non-interactive or minimally interactive computations with tight bounds on communication, error, and privacy. I am applying for this grant to broaden my experience in applied systems and AI privacy through a postdoc at the Max Planck Institute for Security and Privacy, hosted by Carmela Troncoso and Tamalika Mukherjee, after which I will return to Aarhus University for one year to secure an academic position in Denmark.