Accuracy and Privacy Evaluations of Collaborative Data Analysis
This work addresses privacy-preserving distributed data analysis for applications requiring secure collaboration, presenting a novel framework with theoretical guarantees.
The paper tackled the problem of ensuring both accuracy and privacy in collaborative data analysis by analyzing a framework that shares dimensionality-reduced data representations, proving equivalence to centralized analysis under certain conditions and demonstrating protection against insider and external attacks with a double privacy layer.
Distributed data analysis without revealing the individual data has recently attracted significant attention in several applications. A collaborative data analysis through sharing dimensionality reduced representations of data has been proposed as a non-model sharing-type federated learning. This paper analyzes the accuracy and privacy evaluations of this novel framework. In the accuracy analysis, we provided sufficient conditions for the equivalence of the collaborative data analysis and the centralized analysis with dimensionality reduction. In the privacy analysis, we proved that collaborative users' private datasets are protected with a double privacy layer against insider and external attacking scenarios.