Semi-supervised nonlinear dimensionality reduction

Xin Yang, Haoying Fu, Hongyuan Zha, Jesse Louis Barlow

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

36 Citations (Scopus)

Abstract

The problem of nonlinear dimensionality reduction is considered. We focus on problems where prior information is available, namely, semi-supervised dimensionality reduction. It is shown that basic nonlinear dimensionality reduction algorithms, such as Locally Linear Embedding (LLE), Isometric feature mapping (ISOMAP), and Local Tangent Space Alignment (LTSA), can be modified by taking into account prior information on exact mapping of certain data points. The sensitivity analysis of our algorithms shows that prior information will improve stability of the solution. We also give some insight on what kind of prior information best improves the solution. We demonstrate the usefulness of our algorithm by synthetic and real life examples.

Original languageEnglish (US)
Title of host publicationACM International Conference Proceeding Series - Proceedings of the 23rd International Conference on Machine Learning, ICML 2006
Pages1065-1072
Number of pages8
DOIs
StatePublished - Dec 1 2006
Event23rd International Conference on Machine Learning, ICML 2006 - Pittsburgh, PA, United States
Duration: Jun 25 2006Jun 29 2006

Publication series

NameACM International Conference Proceeding Series
Volume148

Other

Other23rd International Conference on Machine Learning, ICML 2006
CountryUnited States
CityPittsburgh, PA
Period6/25/066/29/06

Fingerprint

Sensitivity analysis

All Science Journal Classification (ASJC) codes

  • Software
  • Human-Computer Interaction
  • Computer Vision and Pattern Recognition
  • Computer Networks and Communications

Cite this

Yang, X., Fu, H., Zha, H., & Barlow, J. L. (2006). Semi-supervised nonlinear dimensionality reduction. In ACM International Conference Proceeding Series - Proceedings of the 23rd International Conference on Machine Learning, ICML 2006 (pp. 1065-1072). (ACM International Conference Proceeding Series; Vol. 148). https://doi.org/10.1145/1143844.1143978
Yang, Xin ; Fu, Haoying ; Zha, Hongyuan ; Barlow, Jesse Louis. / Semi-supervised nonlinear dimensionality reduction. ACM International Conference Proceeding Series - Proceedings of the 23rd International Conference on Machine Learning, ICML 2006. 2006. pp. 1065-1072 (ACM International Conference Proceeding Series).
@inproceedings{c8b3555d9f984de6a490c7657d0ef83c,
title = "Semi-supervised nonlinear dimensionality reduction",
abstract = "The problem of nonlinear dimensionality reduction is considered. We focus on problems where prior information is available, namely, semi-supervised dimensionality reduction. It is shown that basic nonlinear dimensionality reduction algorithms, such as Locally Linear Embedding (LLE), Isometric feature mapping (ISOMAP), and Local Tangent Space Alignment (LTSA), can be modified by taking into account prior information on exact mapping of certain data points. The sensitivity analysis of our algorithms shows that prior information will improve stability of the solution. We also give some insight on what kind of prior information best improves the solution. We demonstrate the usefulness of our algorithm by synthetic and real life examples.",
author = "Xin Yang and Haoying Fu and Hongyuan Zha and Barlow, {Jesse Louis}",
year = "2006",
month = "12",
day = "1",
doi = "10.1145/1143844.1143978",
language = "English (US)",
isbn = "1595933832",
series = "ACM International Conference Proceeding Series",
pages = "1065--1072",
booktitle = "ACM International Conference Proceeding Series - Proceedings of the 23rd International Conference on Machine Learning, ICML 2006",

}

Yang, X, Fu, H, Zha, H & Barlow, JL 2006, Semi-supervised nonlinear dimensionality reduction. in ACM International Conference Proceeding Series - Proceedings of the 23rd International Conference on Machine Learning, ICML 2006. ACM International Conference Proceeding Series, vol. 148, pp. 1065-1072, 23rd International Conference on Machine Learning, ICML 2006, Pittsburgh, PA, United States, 6/25/06. https://doi.org/10.1145/1143844.1143978

Semi-supervised nonlinear dimensionality reduction. / Yang, Xin; Fu, Haoying; Zha, Hongyuan; Barlow, Jesse Louis.

ACM International Conference Proceeding Series - Proceedings of the 23rd International Conference on Machine Learning, ICML 2006. 2006. p. 1065-1072 (ACM International Conference Proceeding Series; Vol. 148).

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

TY - GEN

T1 - Semi-supervised nonlinear dimensionality reduction

AU - Yang, Xin

AU - Fu, Haoying

AU - Zha, Hongyuan

AU - Barlow, Jesse Louis

PY - 2006/12/1

Y1 - 2006/12/1

N2 - The problem of nonlinear dimensionality reduction is considered. We focus on problems where prior information is available, namely, semi-supervised dimensionality reduction. It is shown that basic nonlinear dimensionality reduction algorithms, such as Locally Linear Embedding (LLE), Isometric feature mapping (ISOMAP), and Local Tangent Space Alignment (LTSA), can be modified by taking into account prior information on exact mapping of certain data points. The sensitivity analysis of our algorithms shows that prior information will improve stability of the solution. We also give some insight on what kind of prior information best improves the solution. We demonstrate the usefulness of our algorithm by synthetic and real life examples.

AB - The problem of nonlinear dimensionality reduction is considered. We focus on problems where prior information is available, namely, semi-supervised dimensionality reduction. It is shown that basic nonlinear dimensionality reduction algorithms, such as Locally Linear Embedding (LLE), Isometric feature mapping (ISOMAP), and Local Tangent Space Alignment (LTSA), can be modified by taking into account prior information on exact mapping of certain data points. The sensitivity analysis of our algorithms shows that prior information will improve stability of the solution. We also give some insight on what kind of prior information best improves the solution. We demonstrate the usefulness of our algorithm by synthetic and real life examples.

UR - http://www.scopus.com/inward/record.url?scp=34250705299&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34250705299&partnerID=8YFLogxK

U2 - 10.1145/1143844.1143978

DO - 10.1145/1143844.1143978

M3 - Conference contribution

SN - 1595933832

SN - 9781595933836

T3 - ACM International Conference Proceeding Series

SP - 1065

EP - 1072

BT - ACM International Conference Proceeding Series - Proceedings of the 23rd International Conference on Machine Learning, ICML 2006

ER -

Yang X, Fu H, Zha H, Barlow JL. Semi-supervised nonlinear dimensionality reduction. In ACM International Conference Proceeding Series - Proceedings of the 23rd International Conference on Machine Learning, ICML 2006. 2006. p. 1065-1072. (ACM International Conference Proceeding Series). https://doi.org/10.1145/1143844.1143978