ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2306.06723
90
14
v1v2v3 (latest)

Counting Distinct Elements in the Turnstile Model with Differential Privacy under Continual Observation

11 June 2023
Palak Jain
Iden Kalemaj
Sofya Raskhodnikova
Satchit Sivakumar
Adam D. Smith
ArXiv (abs)PDFHTML
Abstract

Privacy is a central challenge for systems that learn from sensitive data sets, especially when a system's outputs must be continuously updated to reflect changing data. We consider the achievable error for differentially private continual release of a basic statistic -- the number of distinct items -- in a stream where items may be both inserted and deleted (the turnstile model). With only insertions, existing algorithms have additive error just polylogarithmic in the length of the stream TTT. We uncover a much richer landscape in the turnstile model, even without considering memory restrictions. We show that every differentially private mechanism that handles insertions and deletions has worst-case additive error at least T1/4T^{1/4}T1/4 even under a relatively weak, event-level privacy definition. Then, we identify a parameter of the input stream, its maximum flippancy, that is low for natural data streams and for which we give tight parameterized error guarantees. Specifically, the maximum flippancy is the largest number of times that the contribution of a single item to the distinct elements count changes over the course of the stream. We present an item-level differentially private mechanism that, for all turnstile streams with maximum flippancy www, continually outputs the number of distinct elements with an O(w⋅polylog⁡T)O(\sqrt{w} \cdot poly\log T)O(w​⋅polylogT) additive error, without requiring prior knowledge of www. We prove that this is the best achievable error bound that depends only on www, for a large range of values of www. When www is small, the error of our mechanism is similar to the polylogarithmic in TTT error in the insertion-only setting, bypassing the hardness in the turnstile model.

View on arXiv
Comments on this paper