Optimality of Frequency Moment Estimation

Mark Braverman, Or Zamir

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Estimating the second frequency moment of a stream up to (1±ϵ) multiplicative error requires at most O(logn / ϵ2) bits of space, due to a seminal result of Alon, Matias, and Szegedy. It is also known that at least ω(logn + 1/ϵ2) space is needed. We prove a tight lower bound of ω(log(n ϵ2 ) / ϵ2) for all ϵ = ω(1/√n). Note that when ϵ>n-1/2 + c, where c>0, our lower bound matches the classic upper bound of AMS. For smaller values of ϵ we also introduce a revised algorithm that improves the classic AMS bound and matches our lower bound. Our lower bound holds also for the more general problem of p-th frequency moment estimation for the range of pϵ (1,2], giving a tight bound in the only remaining range to settle the optimal space complexity of estimating frequency moments.

Original languageEnglish (US)
Title of host publicationSTOC 2025 - Proceedings of the 57th Annual ACM Symposium on Theory of Computing
EditorsMichal Koucky, Nikhil Bansal
PublisherAssociation for Computing Machinery
Pages360-370
Number of pages11
ISBN (Electronic)9798400715105
DOIs
StatePublished - Jun 15 2025
Event57th Annual ACM Symposium on Theory of Computing, STOC 2025 - Prague, Czech Republic
Duration: Jun 23 2025Jun 27 2025

Publication series

NameProceedings of the Annual ACM Symposium on Theory of Computing
ISSN (Print)0737-8017

Conference

Conference57th Annual ACM Symposium on Theory of Computing, STOC 2025
Country/TerritoryCzech Republic
CityPrague
Period6/23/256/27/25

All Science Journal Classification (ASJC) codes

  • Software

Keywords

  • Communication Complexity
  • Frequency Moments
  • Information Theory
  • Streaming Algorithms

Fingerprint

Dive into the research topics of 'Optimality of Frequency Moment Estimation'. Together they form a unique fingerprint.

Cite this