Differential Privacy for Growing Databases

TL;DR

This paper introduces methods for maintaining differential privacy and accuracy in data analysis as databases grow dynamically, enabling private analysis without significant loss of utility over unbounded data accumulation.

Contribution

It presents a general technique and specific algorithms for adapting static differentially private algorithms to dynamic, growing databases, ensuring privacy and accuracy.

Findings

Algorithms can be rerun at data growth points with minimal accuracy loss.

The private multiplicative weights algorithm is adapted for unbounded data growth.

Extensions of several private algorithms to dynamic settings are developed.

Abstract

We study the design of differentially private algorithms for adaptive analysis of dynamically growing databases, where a database accumulates new data entries while the analysis is ongoing. We provide a collection of tools for machine learning and other types of data analysis that guarantee differential privacy and accuracy as the underlying databases grow arbitrarily large. We give both a general technique and a specific algorithm for adaptive analysis of dynamically growing databases. Our general technique is illustrated by two algorithms that schedule black box access to some algorithm that operates on a fixed database to generically transform private and accurate algorithms for static databases into private and accurate algorithms for dynamically growing databases. These results show that almost any private and accurate algorithm can be rerun at appropriate points of data growth with minimal loss of accuracy, even when data growth is unbounded. Our specific algorithm directly adapts the private multiplicative weights algorithm to the dynamic setting, maintaining the accuracy guarantee of the static setting through unbounded data growth. Along the way, we develop extensions of several other differentially private algorithms to the dynamic setting, which may be of independent interest for future work on the design of differentially private algorithms for growing databases.