Single-message shuffled-model frequency estimation has optimal error about min(n^{1/4}, sqrt(B)); multi-message protocols achieve polylogarithmic error with polylogarithmic communication.
Improved Summation from Shuffling
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
A protocol by Ishai et al.\ (FOCS 2006) showing how to implement distributed $n$-party summation from secure shuffling has regained relevance in the context of the recently proposed \emph{shuffle model} of differential privacy, as it allows to attain the accuracy levels of the curator model at a moderate communication cost. To achieve statistical security $2^{-\sigma}$, the protocol by Ishai et al.\ requires the number of messages sent by each party to {\em grow} logarithmically with $n$ as $O(\log n + \sigma)$. In this note we give an improved analysis achieving a dependency of the form $O(1+\sigma/\log n)$. Conceptually, this addresses the intuitive question left open by Ishai et al.\ of whether the shuffling step in their protocol provides a "hiding in the crowd" amplification effect as $n$ increases. From a practical perspective, our analysis provides explicit constants and shows, for example, that the method of Ishai et al.\ applied to summation of $32$-bit numbers from $n=10^4$ parties sending $12$ messages each provides statistical security $2^{-40}$.
fields
cs.CR 1years
2019 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
On the Power of Multiple Anonymous Messages
Single-message shuffled-model frequency estimation has optimal error about min(n^{1/4}, sqrt(B)); multi-message protocols achieve polylogarithmic error with polylogarithmic communication.