Linear hash functions

Author:

Alon Noga1,Dietzfelbinger Martin2,Miltersen Peter Bro3,Petrank Erez4,Tardos Gábor5

Affiliation:

1. Tel-Aviv Univ., Tel-Aviv, Israel; and Institute for Advanced Study, Princeton, NJ

2. Technische Univ. Ilmenau, Ilmenau, Germany

3. Univ. of Aarhus, Aarhus, Denmark

4. IBM, Haifa, Israel

5. Hugarian Academy of Sciences, Budapest, Hungary

Abstract

Consider the set ℋ of all linear (or affine) transformations between two vector spaces over a finite field F . We study how good ℋ is as a class of hash functions, namely we consider hashing a set S of size n into a range having the same cardinality n by a randomly chosen function from ℋ and look at the expected size of the largest hash bucket. ℋ is a universal class of hash functions for any finite field, but with respect to our measure different fields behave differently. If the finite field F has n elements, then there is a bad set SF 2 of size n with expected maximal bucket size Ω( n 1/3 ). If n is a perfect square, then there is even a bad set with largest bucket size always at least √n. (This is worst possible, since with respect to a universal class of hash functions every set of size n has expected largest bucket size below √ + 1/2.) If, however, we consider the field of two elements, then we get much better bounds. The best previously known upper bound on the expected size of the largest bucket for this class was O (2 √ log n ). We reduce this upper bound to O (log n log log n ). Note that this is not far from the guarantee for a random function. There, the average largest bucket would be Θ (log n / log log n ). In the course of our proof we develop a tool which may be of independent interest. Suppose we have a subset S of a vector space D over Z 2 , and consider a random linear mapping of D to a smaller vector space R . If the cardinality of S is larger than c ε | R |log| R |, then with probability 1 - ϵ, the image of S will cover all elements in the range.

Publisher

Association for Computing Machinery (ACM)

Subject

Artificial Intelligence,Hardware and Architecture,Information Systems,Control and Systems Engineering,Software

Reference17 articles.

1. A fast and simple randomized parallel algorithm for the maximal independent set problem

2. Universal classes of hash functions;CARTER J.L.;J. Comput. Syst. Sci.,1979

3. Lecture Notes in Computer Science;DIETZFELBINGER M.

Cited by 19 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Linear Hashing with ℓ guarantees and two-sided Kakeya bounds;2022 IEEE 63rd Annual Symposium on Foundations of Computer Science (FOCS);2022-10

2. HalftimeHash: Modern Hashing Without 64-Bit Multipliers or Finite Fields;Lecture Notes in Computer Science;2021

3. Public-Coin Statistical Zero-Knowledge Batch Verification Against Malicious Verifiers;Lecture Notes in Computer Science;2021

4. Extractors and Secret Sharing Against Bounded Collusion Protocols;2020 IEEE 61st Annual Symposium on Foundations of Computer Science (FOCS);2020-11

5. Linear Hashing Is Awesome;SIAM Journal on Computing;2019-01

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3