Multiparty Differential Privacy via Aggregation of Locally Trained Classifiers

Abstract

As increasing amounts of sensitive personal information finds its way into data repositories, it is important to develop analysis mechanisms that can derive ag-gregate information from these repositories without revealing information about individual data instances. Though the differential privacy model provides a frame-work to analyze such mechanisms for databases belonging to a single party, this framework has not yet been considered in a multi-party setting. In this paper, we propose a privacy-preserving protocol for composing a differentially private ag-gregate classifier using classifiers trained locally by separate mutually untrusting parties. The protocol allows these parties to interact with an untrusted curator to construct additive shares of a perturbed aggregate classifier. We also present a detailed theoretical analysis containing a proof of differential privacy of the per-turbed aggregate classifier and a bound on the excess risk introduced by the per-turbation. We verify the bound with an experimental evaluation on a real dataset. 1