Differentially Private Summation with Multi-Message Shuffling

Abstract

In recent work, Cheu et al. (Eurocrypt 2019) proposed a protocol for $n$-party real summation in the shuffle model of differential privacy with $O_{\epsilon ,\delta }(1)$ error and $\Theta (\epsilon \sqrt{n})$ one-bit messages per party. In contrast, every local model protocol for real summation must incur error $\Omega (1/\sqrt{n})$, and there exist protocols matching this lower bound which require just one bit of communication per party. Whether this gap in number of messages is necessary was left open by Cheu et al. In this note we show a protocol with $O(1/\epsilon )$ error and $O(\log (n/\delta ))$ messages of size $O(\log (n))$ per party. This protocol is based on the work of Ishai et al.\ (FOCS 2006) showing how to implement distributed summation from secure shuffling, and the observation that this allows simulating the Laplace mechanism in the shuffle model.

References in corpus (2)

• Privacy for Free: Posterior Sampling and Stochastic Gradient Monte Carlo
• Scalable and Differentially Private Distributed Aggregation in the Shuffled Model

Cited by in corpus (10)

• Scalable and Differentially Private Distributed Aggregation in the Shuffled Model
• Encode, Shuffle, Analyze Privacy Revisited: Formalizations and Empirical Evaluation
• On the Power of Multiple Anonymous Messages
• Improved Summation from Shuffling
• Shuffled Model of Federated Learning: Privacy, Communication and Accuracy Trade-offs
• Connecting Robust Shuffle Privacy and Pan-Privacy
• Renyi Differential Privacy of the Subsampled Shuffle Model in Distributed Learning
• Local Differential Privacy for Regret Minimization in Reinforcement Learning
• Differential Privacy in the Shuffle Model: A Survey of Separations
• Privacy Enhancement via Dummy Points in the Shuffle Model