CMI Seminar: Fedor Sandomirskiy (part 2 of 2)

In a private private information structure, agents' signals contain no information about the signals of their peers. We study how informative such structures can be and characterize those that are Pareto optimal, in the sense that it is impossible to add information to any of the agents without violating privacy.

In the first part of the talk, we will discuss structural results and information-theoretic limitations to informativeness. In the second part, we will describe the Pareto frontier using an (unexpected) connection to mathematical tomography.

Joint work with Kevin He (UPenn) and Omer Tamuz (Caltech).