Practical data-dependent metric compression with provable guarantees
Name
6855-practical-data-dependent-metric-compression-with-provable-guarantees.pdf
Description
Published version
Size
586.68 KB
Format
Adobe PDF
Checksum (MD5)
a1f2d2bda154ea59613e7b5adc087ba2
Author(s)
Indyk, Piotr
Razenshteyn, Ilya
Wagner, Tal
Date Issued
December 2017
Journal
31st Conference on Neural Information Processing Systems (NIPS 2017)
Publisher
Neural Information Processing Systems Foundation, Inc.
Citation
Indyk, Piotr, Ilya Razenshteyn, and Tal Wagner. "Practical Data-Dependent Metric Compression with
Provable Guarantees." 31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, California, USA, 4-9 December, 2017, NIPS, 2017. © Neural Information Processing Systems Foundation, Inc.
Version
Final published version
Abstract
We introduce a new distance-preserving compact representation of multidimensional point-sets. Given n points in a d-dimensional space where each coordinate is represented using B bits (i.e., dB bits per point), it produces a representation of size O(dlog(dB/e) + logn) bits per point from which one can approximate the distances up to a factor of 1 ± e. Our algorithm almost matches the recent bound of [6] while being much simpler. We compare our algorithm to Product Quantization (PQ) [7], a state of the art heuristic metric compression method. We evaluate both algorithms on several data sets: SIFT (used in [7]), MNIST [11], New York City taxi time series [4] and a synthetic one-dimensional data set embedded in a high-dimensional space. With appropriately tuned parameters, our algorithm produces representations that are comparable to or better than those produced by PQ, while having provable guarantees on its performance.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
https://papers.nips.cc/paper/6855-practical-data-dependent-metric-compression-with-provable-guarantees
