Learning nonlinear manifolds from time series

Ruei Sung Lin, Che Bin Liu, Ming Hsuan Yang, Narendra Ahuja, Stephen E Levinson

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

Abstract

There has been growing interest in developing nonlinear dimensionality reduction algorithms for vision applications. Although progress has been made in recent years, conventional nonlinear dimensionality reduction algorithms have been designed to deal with stationary, or independent and identically distributed data. In this paper, we present a novel method that learns nonlinear mapping from time series data to their intrinsic coordinates on the underlying manifold. Our work extends the recent advances in learning nonlinear manifolds within a global coordinate system to account for temporal correlation inherent in sequential data. We formulate the problem with a dynamic Bayesian network and propose an approximate algorithm to tackle the learning and inference problems. Numerous experiments demonstrate the proposed method is able to learn nonlinear manifolds from time series data, and as a result of exploiting the temporal correlation, achieve superior results.

Original languageEnglish (US)
Title of host publicationComputer Vision - ECCV 2006, 9th European Conference on Computer Vision, Proceedings
Pages245-256
Number of pages12
StatePublished - Jul 17 2006
Event9th European Conference on Computer Vision, ECCV 2006 - Graz, Austria
Duration: May 7 2006May 13 2006

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3952 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other9th European Conference on Computer Vision, ECCV 2006
CountryAustria
CityGraz
Period5/7/065/13/06

Fingerprint

Time series
Temporal Correlation
Dimensionality Reduction
Time Series Data
Bayesian networks
Dynamic Bayesian Networks
Nonlinear Mapping
Approximate Algorithm
Identically distributed
Learning
Experiments
Demonstrate
Experiment

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Lin, R. S., Liu, C. B., Yang, M. H., Ahuja, N., & Levinson, S. E. (2006). Learning nonlinear manifolds from time series. In Computer Vision - ECCV 2006, 9th European Conference on Computer Vision, Proceedings (pp. 245-256). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3952 LNCS).

Learning nonlinear manifolds from time series. / Lin, Ruei Sung; Liu, Che Bin; Yang, Ming Hsuan; Ahuja, Narendra; Levinson, Stephen E.

Computer Vision - ECCV 2006, 9th European Conference on Computer Vision, Proceedings. 2006. p. 245-256 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3952 LNCS).

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

Lin, RS, Liu, CB, Yang, MH, Ahuja, N & Levinson, SE 2006, Learning nonlinear manifolds from time series. in Computer Vision - ECCV 2006, 9th European Conference on Computer Vision, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 3952 LNCS, pp. 245-256, 9th European Conference on Computer Vision, ECCV 2006, Graz, Austria, 5/7/06.
Lin RS, Liu CB, Yang MH, Ahuja N, Levinson SE. Learning nonlinear manifolds from time series. In Computer Vision - ECCV 2006, 9th European Conference on Computer Vision, Proceedings. 2006. p. 245-256. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Lin, Ruei Sung ; Liu, Che Bin ; Yang, Ming Hsuan ; Ahuja, Narendra ; Levinson, Stephen E. / Learning nonlinear manifolds from time series. Computer Vision - ECCV 2006, 9th European Conference on Computer Vision, Proceedings. 2006. pp. 245-256 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{e6c0f192ce9e45b69366557be6eb478c,
title = "Learning nonlinear manifolds from time series",
abstract = "There has been growing interest in developing nonlinear dimensionality reduction algorithms for vision applications. Although progress has been made in recent years, conventional nonlinear dimensionality reduction algorithms have been designed to deal with stationary, or independent and identically distributed data. In this paper, we present a novel method that learns nonlinear mapping from time series data to their intrinsic coordinates on the underlying manifold. Our work extends the recent advances in learning nonlinear manifolds within a global coordinate system to account for temporal correlation inherent in sequential data. We formulate the problem with a dynamic Bayesian network and propose an approximate algorithm to tackle the learning and inference problems. Numerous experiments demonstrate the proposed method is able to learn nonlinear manifolds from time series data, and as a result of exploiting the temporal correlation, achieve superior results.",
author = "Lin, {Ruei Sung} and Liu, {Che Bin} and Yang, {Ming Hsuan} and Narendra Ahuja and Levinson, {Stephen E}",
year = "2006",
month = "7",
day = "17",
language = "English (US)",
isbn = "3540338349",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "245--256",
booktitle = "Computer Vision - ECCV 2006, 9th European Conference on Computer Vision, Proceedings",

}

TY - GEN

T1 - Learning nonlinear manifolds from time series

AU - Lin, Ruei Sung

AU - Liu, Che Bin

AU - Yang, Ming Hsuan

AU - Ahuja, Narendra

AU - Levinson, Stephen E

PY - 2006/7/17

Y1 - 2006/7/17

N2 - There has been growing interest in developing nonlinear dimensionality reduction algorithms for vision applications. Although progress has been made in recent years, conventional nonlinear dimensionality reduction algorithms have been designed to deal with stationary, or independent and identically distributed data. In this paper, we present a novel method that learns nonlinear mapping from time series data to their intrinsic coordinates on the underlying manifold. Our work extends the recent advances in learning nonlinear manifolds within a global coordinate system to account for temporal correlation inherent in sequential data. We formulate the problem with a dynamic Bayesian network and propose an approximate algorithm to tackle the learning and inference problems. Numerous experiments demonstrate the proposed method is able to learn nonlinear manifolds from time series data, and as a result of exploiting the temporal correlation, achieve superior results.

AB - There has been growing interest in developing nonlinear dimensionality reduction algorithms for vision applications. Although progress has been made in recent years, conventional nonlinear dimensionality reduction algorithms have been designed to deal with stationary, or independent and identically distributed data. In this paper, we present a novel method that learns nonlinear mapping from time series data to their intrinsic coordinates on the underlying manifold. Our work extends the recent advances in learning nonlinear manifolds within a global coordinate system to account for temporal correlation inherent in sequential data. We formulate the problem with a dynamic Bayesian network and propose an approximate algorithm to tackle the learning and inference problems. Numerous experiments demonstrate the proposed method is able to learn nonlinear manifolds from time series data, and as a result of exploiting the temporal correlation, achieve superior results.

UR - http://www.scopus.com/inward/record.url?scp=33745847556&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33745847556&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:33745847556

SN - 3540338349

SN - 9783540338345

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 245

EP - 256

BT - Computer Vision - ECCV 2006, 9th European Conference on Computer Vision, Proceedings

ER -