On the impossibility of dimension reduction in l 1

Author:

Brinkman Bo1,Charikar Moses2

Affiliation:

1. Miami University, Oxford, Ohio

2. Princeton University, Princeton, New Jersey

Abstract

The Johnson--Lindenstrauss lemma shows that any n points in Euclidean space (i.e., ℝ n with distances measured under the ℓ 2 norm) may be mapped down to O ((log n )/ϵ 2 ) dimensions such that no pairwise distance is distorted by more than a (1 + ϵ) factor. Determining whether such dimension reduction is possible in ℓ 1 has been an intriguing open question. We show strong lower bounds for general dimension reduction in ℓ 1 . We give an explicit family of n points in ℓ 1 such that any embedding with constant distortion D requires n Ω(1/ D 2 ) dimensions. This proves that there is no analog of the Johnson--Lindenstrauss lemma for ℓ 1 ; in fact, embedding with any constant distortion requires n Ω(1) dimensions. Further, embedding the points into ℓ 1 with (1+ϵ) distortion requires n ½− O (ϵ log(1/ϵ)) dimensions. Our proof establishes this lower bound for shortest path metrics of series-parallel graphs. We make extensive use of linear programming and duality in devising our bounds. We expect that the tools and techniques we develop will be useful for future investigations of embeddings into ℓ 1 .

Publisher

Association for Computing Machinery (ACM)

Subject

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

Reference29 articles.

Cited by 52 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. Covering Planar Metrics (and Beyond): O(1) Trees Suffice;2023 IEEE 64th Annual Symposium on Foundations of Computer Science (FOCS);2023-11-06

2. $$\varepsilon $$-Isometric Dimension Reduction for Incompressible Subsets of $$\ell _p$$;Discrete & Computational Geometry;2023-10-21

3. Matrix deviation inequality for ℓp-norm;Random Matrices: Theory and Applications;2023-06-15

4. Exposé Bourbaki 1188 : Average distortion embeddings, nonlinear spectral gaps, and a metric John theorem [after Assaf Naor];Astérisque;2023-04-04

5. Random Projection;Elements of Dimensionality Reduction and Manifold Learning;2023

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