English

Taylor Polynomial Estimator for Estimating Frequency Moments

Data Structures and Algorithms 2015-06-05 v1

Abstract

We present a randomized algorithm for estimating the ppth moment FpF_p of the frequency vector of a data stream in the general update (turnstile) model to within a multiplicative factor of 1±ϵ1 \pm \epsilon, for p>2p > 2, with high constant confidence. For 0<ϵ10 < \epsilon \le 1, the algorithm uses space O(n12/pϵ2+n12/pϵ4/plog(n))O( n^{1-2/p} \epsilon^{-2} + n^{1-2/p} \epsilon^{-4/p} \log (n)) words. This improves over the current bound of O(n12/pϵ24/plog(n))O(n^{1-2/p} \epsilon^{-2-4/p} \log (n)) words by Andoni et. al. in \cite{ako:arxiv10}. Our space upper bound matches the lower bound of Li and Woodruff \cite{liwood:random13} for ϵ=(log(n))Ω(1)\epsilon = (\log (n))^{-\Omega(1)} and the lower bound of Andoni et. al. \cite{anpw:icalp13} for ϵ=Ω(1)\epsilon = \Omega(1).

Keywords

Cite

@article{arxiv.1506.01442,
  title  = {Taylor Polynomial Estimator for Estimating Frequency Moments},
  author = {Sumit Ganguly},
  journal= {arXiv preprint arXiv:1506.01442},
  year   = {2015}
}

Comments

Supercedes arXiv:1104.4552. Extended Abstract of this paper to appear in Proceedings of ICALP 2015

R2 v1 2026-06-22T09:47:00.465Z