arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1201.0253cs.DScs.CC
收藏

数据流高阶矩估计的一个下界

A Lower Bound for Estimating High Moments of a Data Stream

Sumit Ganguly

摘要:

我们显示了一个改进的下界的Fp估计问题的数据流设置为p>2。数据流是来自域[n]的具有可能重复的项的序列。频率向量x是n维非负整数向量x,使得x(i)是i在序列中出现的次数。给定精度参数Omega(n^{-1/p})< ε< 1,估计Fp的问题是以在线方式并使用尽可能少的空间以高恒定概率正确地估计\norm{x}_p^p = \sum_{i \in [n]} \abs{x(i)}^p到1 \pm ε的相对精度内。该问题的当前空间下界为Ω(n^{1-2/p} ε^{-2/p}+ n^{1-2/p}ε^{-4/p}/ \log^{O(1)}(n)+(ε^{-2} + \log(n)。下界表达式中的第一项在\cite{B-YJKS:stoc02,cks:ccc03}中得到证明,第二项在\cite{wz:arxiv11}中得到证明,第三项在\cite{wood:soda04}中得到证明。在本说明中,我们展示了Omega(p^2 n^{1-2/p} ε^{-2}/ \log(n))位空间界限,其中Omega(pn^{-1/p})\le ε \le 1/10。

英文摘要:

We show an improved lower bound for the Fp estimation problem in a data stream setting for p>2. A data stream is a sequence of items from the domain [n] with possible repetitions. The frequency vector x is an n-dimensional non-negative integer vector x such that x(i) is the number of occurrences of i in the sequence. Given an accuracy parameter Omega(n^{-1/p}) < ε< 1, the problem of estimating Fp is to estimate \norm{x}_p^p = \sum_{i \in [n]} \abs{x(i)}^p correctly to within a relative accuracy of 1\pm εwith high constant probability in an online fashion and using as little space as possible. The current space lower bound for this problem is Omega(n^{1-2/p} ε^{-2/p}+ n^{1-2/p}ε^{-4/p}/ \log^{O(1)}(n)+ (ε^{-2} + \log (n))). The first term in the lower bound expression was proved in \cite{B-YJKS:stoc02,cks:ccc03}, the second in \cite{wz:arxiv11} and the third in \cite{wood:soda04}. In this note, we show an Omega(p^2 n^{1-2/p} ε^{-2}/\log (n)) bits space bound, for Omega(pn^{-1/p}) \le ε\le 1/10.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑