Private Federated Statistics in an Interactive Setting

11/18/2022

∙

Privately learning statistics of events on devices can enable improved user experience. Differentially private algorithms for such problems can benefit significantly from interactivity. We argue that an aggregation protocol can enable an interactive private federated statistics system where user's devices maintain control of the privacy assurance. We describe the architecture of such a system, and analyze its security properties.

READ FULL TEXT

Private Federated Statistics in an Interactive Setting

Efficient Differentially Private Secure Aggregation for Federated Learning via Hardness of Learning with Errors

Samplable Anonymous Aggregation for Private Federated Data Analysis

A Primer on Private Statistics

Federated and Differentially Private Learning for Electronic Health Records

Private Multi-Group Aggregation

Private Hierarchical Clustering in Federated Networks

The Target-Charging Technique for Privacy Accounting across Interactive Computations

Private Federated Statistics in an Interactive Setting

Related Research

Efficient Differentially Private Secure Aggregation for Federated Learning via Hardness of Learning with Errors

Samplable Anonymous Aggregation for Private Federated Data Analysis

A Primer on Private Statistics

Federated and Differentially Private Learning for Electronic Health Records

Private Multi-Group Aggregation

Private Hierarchical Clustering in Federated Networks

The Target-Charging Technique for Privacy Accounting across Interactive Computations