Privacy-preserving Data Sharing on Vertically Partitioned Data

10/19/2020
by   Razane Tajeddine, et al.
0

In this work, we present a method for differentially private data sharing by training a mixture model on vertically partitioned data, where each party holds different features for the same set of individuals. We use secure multi-party computation (MPC) to combine the contribution of the data from the parties to train the model. We apply the differentially private variational inference (DPVI) for learning the model. Assuming the mixture components contain no dependencies across different parties, the objective function can be factorized into a sum of products of individual components of each party. Therefore, each party can calculate its shares on its own without the use of MPC. Then MPC is only needed to get the product between the different shares and add the noise. Applying the method to demographic data from the US Census, we obtain comparable accuracy to the non-partitioned case with approximately 20-fold increase in computing time.

READ FULL TEXT
research
09/06/2020

Hybrid Differentially Private Federated Learning on Vertically Partitioned Data

We present HDP-VFL, the first hybrid differentially private (DP) framewo...
research
09/23/2022

Differentially private partitioned variational inference

Learning a privacy-preserving model from distributed sensitive data is a...
research
10/20/2021

Differential Privacy in Multi-Party Resource Sharing

This study examines a resource-sharing problem involving multiple partie...
research
04/15/2021

Fast Private Parameter Learning and Evaluation for Sum-Product Networks

A sum-product network (SPN) is a graphical model that allows several typ...
research
08/07/2022

CoVault: A Secure Analytics Platform

In a secure analytics platform, data sources consent to the exclusive us...
research
07/26/2021

Selective MPC: Distributed Computation of Differentially Private Key Value Statistics

An increasingly popular method for computing aggregate statistics while ...
research
12/30/2021

Circuit-Free General-Purpose Multi-Party Computation via Co-Utile Unlinkable Outsourcing

Multiparty computation (MPC) consists in several parties engaging in joi...

Please sign up or login with your details

Forgot password? Click here to reset