Exponential Separations in Local Differential Privacy
Matthew Joseph, Jieming Mao, Aaron Roth
摘要
We prove a general connection between the communication complexity of two-player games and the sample complexity of their multi-player locally private analogues. We use this connection to prove sample complexity lower bounds for locally differentially private protocols as straightforward corollaries of results from communication complexity. In particular, we 1) use a communication lower bound for the hidden layers problem to prove an exponential sample complexity separation between sequentially and fully interactive locally private protocols, and 2) use a communication lower bound for the pointer chasing problem to prove an exponential sample complexity separation between k-round and (k + 1)-round sequentially interactive locally private protocols, for every k.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper7
- Locally Differentially Private Analysis of Graph StatisticsJacob Imola, Takao Murakami, Kamalika ChaudhuriUSENIX Security 2021 · 被引用 139 次
- AHEAD: Adaptive Hierarchical Decomposition for Range Query under Local Differential PrivacyLinkang Du, Zhikun Zhang, Shaojie Bai, Changchang Liu 等CCS 2021 · 被引用 32 次
- Connecting Robust Shuffle Privacy and Pan-PrivacyVictor Balcer, Albert Cheu, Matthew Joseph, Jieming MaoSODA 2021 · 被引用 27 次
- Locally differentially private estimation of functionals of discrete distributionsCristina Butucea, Yann IssartelNeurIPS 2021 · 被引用 9 次
- Interaction is necessary for distributed learning with privacy or communication constraintsYuval Dagan, Vitaly FeldmanSTOC 2020 · 被引用 1 次
它引用的顶会 Paper1
相关 Paper
- Pointwise Bounds for Distribution Estimation under Communication ConstraintsWei-Ning Chen, Peter Kairouz, Ayfer ÖzgürNeurIPS 2021 · 被引用 8 次
- Locally Private k-Means Clustering with Constant Multiplicative Approximation and Near-Optimal Additive ErrorAnamay Chaturvedi, Matthew Jones, Huy Le NguyenAAAI 2022 · 被引用 5 次
- Magic and Communication ComplexityUma Girish, Alex May, Natalie Parham, Henry YuenSTOC 2026 · 被引用 5 次
- On Differential Privacy and Adaptive Data Analysis with Bounded SpaceItai Dinur, Uri Stemmer, David P. Woodruff, Samson ZhouEUROCRYPT 2023 · 被引用 5 次
- New separations results for external informationMark Braverman, Dor MinzerSTOC 2021
