Concentrated Differential Privacy

TL;DR

This paper introduces Concentrated Differential Privacy, a new privacy framework that offers improved accuracy over traditional differential privacy methods while maintaining strong privacy guarantees over multiple computations.

Contribution

It presents a novel privacy definition that balances accuracy and privacy, improving upon existing differential privacy relaxations.

Findings

Achieves better accuracy than pure differential privacy.

Maintains privacy guarantees over multiple computations.

Offers a practical alternative to existing privacy frameworks.

Abstract

We introduce Concentrated Differential Privacy, a relaxation of Differential Privacy enjoying better accuracy than both pure differential privacy and its popular "(epsilon,delta)" relaxation without compromising on cumulative privacy loss over multiple computations.