Multiple organizations often wish to aggregate their sensitive data and learn from it, but they cannot do so because they cannot share their data.
For example, healthcare organizations wish to train models jointly over their aggregate patient data to discover better medical diagnosis or treatments, but they often cannot share this data due to privacy concerns or regulations.
To address such problems, my students and I developed MC^2, a framework for secure collaborative computation.
In this talk, I will describe MC^2 and the new possibilities it brings.