Lower bounds on Locality Sensitive Hashing
Rajeev Motwani, Assaf Naor, Rina Panigrahy

TL;DR
This paper establishes a lower bound on the efficiency of locality sensitive hashing in the space, showing that certain optimal parameters cannot be improved beyond a specific threshold, thus guiding future research directions.
Contribution
The paper proves a lower bound on the parameter -LSH, demonstrating that space cannot achieve -sensitive hash families with c parameter less than 1/(2c), nearly matching existing upper bounds.
Findings
Proves -LSH lower bound c c parameter cannot be improved beyond 1/(2c).
Almost matches the upper bound construction by Indyk and Motwani.
Provides theoretical limits guiding the design of locality sensitive hashing algorithms.
Abstract
Given a metric space , , , and , a distribution over mappings is called a -sensitive hash family if any two points in at distance at most are mapped by to the same value with probability at least , and any two points at distance greater than are mapped by to the same value with probability at most . This notion was introduced by Indyk and Motwani in 1998 as the basis for an efficient approximate nearest neighbor search algorithm, and has since been used extensively for this purpose. The performance of these algorithms is governed by the parameter , and constructing hash families with small automatically yields improved nearest neighbor algorithms. Here we show that for it is impossible to achieve . This almost matches…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdvanced Image and Video Retrieval Techniques · Algorithms and Data Compression · Robotics and Sensor-Based Localization
