R2SDH: Robust rotated supervised discrete hashing

Jie Gui, Ping Li

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

Learning-based hashing has recently received considerable attentions due to its capability of supporting efficient storage and retrieval of high-dimensional data such as images, videos, and documents. In this paper, we propose a learning-based hashing algorithm called “Robust Rotated Supervised Discrete Hashing” (R2SDH), by extending the previous work on “Supervised Discrete Hashing” (SDH). In R2SDH, correntropy is adopted to replace the least square regression (LSR) model in SDH for achieving better robustness. Furthermore, considering the commonly used distance metrics such as cosine and Euclidean distance are invariant to rotational transformation, rotation is integrated into the original zero-one label matrix used in SDH, as additional freedom to promote flexibility without sacrificing accuracy. The rotation matrix is learned through an optimization procedure. Experimental results on three image datasets (MNIST, CIFAR-10, and NUS-WIDE) confirm that R2SDH generally outperforms SDH.

Original languageEnglish (US)
Title of host publicationKDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
PublisherAssociation for Computing Machinery
Pages1485-1493
Number of pages9
ISBN (Print)9781450355520
DOIs
StatePublished - Jul 19 2018
Event24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018 - London, United Kingdom
Duration: Aug 19 2018Aug 23 2018

Publication series

NameProceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Other

Other24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018
CountryUnited Kingdom
CityLondon
Period8/19/188/23/18

Fingerprint

Labels

All Science Journal Classification (ASJC) codes

  • Software
  • Information Systems

Keywords

  • Robust M-estimator
  • Rotation
  • Supervised discrete hashing

Cite this

Gui, J., & Li, P. (2018). R2SDH: Robust rotated supervised discrete hashing. In KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1485-1493). (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). Association for Computing Machinery. https://doi.org/10.1145/3219819.3219955
Gui, Jie ; Li, Ping. / R2SDH : Robust rotated supervised discrete hashing. KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2018. pp. 1485-1493 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).
@inproceedings{c96f463a079d4e19a3fb3a1ac24613fc,
title = "R2SDH: Robust rotated supervised discrete hashing",
abstract = "Learning-based hashing has recently received considerable attentions due to its capability of supporting efficient storage and retrieval of high-dimensional data such as images, videos, and documents. In this paper, we propose a learning-based hashing algorithm called “Robust Rotated Supervised Discrete Hashing” (R2SDH), by extending the previous work on “Supervised Discrete Hashing” (SDH). In R2SDH, correntropy is adopted to replace the least square regression (LSR) model in SDH for achieving better robustness. Furthermore, considering the commonly used distance metrics such as cosine and Euclidean distance are invariant to rotational transformation, rotation is integrated into the original zero-one label matrix used in SDH, as additional freedom to promote flexibility without sacrificing accuracy. The rotation matrix is learned through an optimization procedure. Experimental results on three image datasets (MNIST, CIFAR-10, and NUS-WIDE) confirm that R2SDH generally outperforms SDH.",
keywords = "Robust M-estimator, Rotation, Supervised discrete hashing",
author = "Jie Gui and Ping Li",
year = "2018",
month = "7",
day = "19",
doi = "10.1145/3219819.3219955",
language = "English (US)",
isbn = "9781450355520",
series = "Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",
publisher = "Association for Computing Machinery",
pages = "1485--1493",
booktitle = "KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

}

Gui, J & Li, P 2018, R2SDH: Robust rotated supervised discrete hashing. in KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Association for Computing Machinery, pp. 1485-1493, 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018, London, United Kingdom, 8/19/18. https://doi.org/10.1145/3219819.3219955

R2SDH : Robust rotated supervised discrete hashing. / Gui, Jie; Li, Ping.

KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2018. p. 1485-1493 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - R2SDH

T2 - Robust rotated supervised discrete hashing

AU - Gui, Jie

AU - Li, Ping

PY - 2018/7/19

Y1 - 2018/7/19

N2 - Learning-based hashing has recently received considerable attentions due to its capability of supporting efficient storage and retrieval of high-dimensional data such as images, videos, and documents. In this paper, we propose a learning-based hashing algorithm called “Robust Rotated Supervised Discrete Hashing” (R2SDH), by extending the previous work on “Supervised Discrete Hashing” (SDH). In R2SDH, correntropy is adopted to replace the least square regression (LSR) model in SDH for achieving better robustness. Furthermore, considering the commonly used distance metrics such as cosine and Euclidean distance are invariant to rotational transformation, rotation is integrated into the original zero-one label matrix used in SDH, as additional freedom to promote flexibility without sacrificing accuracy. The rotation matrix is learned through an optimization procedure. Experimental results on three image datasets (MNIST, CIFAR-10, and NUS-WIDE) confirm that R2SDH generally outperforms SDH.

AB - Learning-based hashing has recently received considerable attentions due to its capability of supporting efficient storage and retrieval of high-dimensional data such as images, videos, and documents. In this paper, we propose a learning-based hashing algorithm called “Robust Rotated Supervised Discrete Hashing” (R2SDH), by extending the previous work on “Supervised Discrete Hashing” (SDH). In R2SDH, correntropy is adopted to replace the least square regression (LSR) model in SDH for achieving better robustness. Furthermore, considering the commonly used distance metrics such as cosine and Euclidean distance are invariant to rotational transformation, rotation is integrated into the original zero-one label matrix used in SDH, as additional freedom to promote flexibility without sacrificing accuracy. The rotation matrix is learned through an optimization procedure. Experimental results on three image datasets (MNIST, CIFAR-10, and NUS-WIDE) confirm that R2SDH generally outperforms SDH.

KW - Robust M-estimator

KW - Rotation

KW - Supervised discrete hashing

UR - http://www.scopus.com/inward/record.url?scp=85051567835&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051567835&partnerID=8YFLogxK

U2 - 10.1145/3219819.3219955

DO - 10.1145/3219819.3219955

M3 - Conference contribution

AN - SCOPUS:85051567835

SN - 9781450355520

T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

SP - 1485

EP - 1493

BT - KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

PB - Association for Computing Machinery

ER -

Gui J, Li P. R2SDH: Robust rotated supervised discrete hashing. In KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery. 2018. p. 1485-1493. (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). https://doi.org/10.1145/3219819.3219955