Abstract

The problem of domain adaptation (DA) deals with adapting classifier models trained on one data distribution to different data distributions. In this paper, we introduce the Nonlinear Embedding Transform (NET) for unsupervised DA by combining domain alignment along with similarity-based embedding. We also introduce a validation procedure to estimate the model parameters for the NET algorithm using the source data. Comprehensive evaluations on multiple vision datasets demonstrate that the NET algorithm outperforms existing competitive procedures for unsupervised DA.

Original languageEnglish (US)
Title of host publicationComputer Vision – ECCV 2016 Workshops, Proceedings
PublisherSpringer Verlag
Pages451-457
Number of pages7
Volume9915 LNCS
ISBN (Print)9783319494081
DOIs
StatePublished - 2016
Event14th European Conference on Computer Vision, ECCV 2016 - Amsterdam, Netherlands
Duration: Oct 8 2016Oct 16 2016

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9915 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other14th European Conference on Computer Vision, ECCV 2016
CountryNetherlands
CityAmsterdam
Period10/8/1610/16/16

Fingerprint

Mathematical transformations
Transform
Data Distribution
Comprehensive Evaluation
Classifiers
Alignment
Classifier
Model
Estimate
Demonstrate
Similarity
Vision

Keywords

  • MMD
  • Nonlinear embedding
  • Unsupervised
  • Validation

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Demakethepalli Venkateswara, H., Chakraborty, S., & Panchanathan, S. (2016). Nonlinear embedding transform for unsupervised domain adaptation. In Computer Vision – ECCV 2016 Workshops, Proceedings (Vol. 9915 LNCS, pp. 451-457). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9915 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-49409-8_36

Nonlinear embedding transform for unsupervised domain adaptation. / Demakethepalli Venkateswara, Hemanth; Chakraborty, Shayok; Panchanathan, Sethuraman.

Computer Vision – ECCV 2016 Workshops, Proceedings. Vol. 9915 LNCS Springer Verlag, 2016. p. 451-457 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9915 LNCS).

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

Demakethepalli Venkateswara, H, Chakraborty, S & Panchanathan, S 2016, Nonlinear embedding transform for unsupervised domain adaptation. in Computer Vision – ECCV 2016 Workshops, Proceedings. vol. 9915 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9915 LNCS, Springer Verlag, pp. 451-457, 14th European Conference on Computer Vision, ECCV 2016, Amsterdam, Netherlands, 10/8/16. https://doi.org/10.1007/978-3-319-49409-8_36
Demakethepalli Venkateswara H, Chakraborty S, Panchanathan S. Nonlinear embedding transform for unsupervised domain adaptation. In Computer Vision – ECCV 2016 Workshops, Proceedings. Vol. 9915 LNCS. Springer Verlag. 2016. p. 451-457. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-49409-8_36
Demakethepalli Venkateswara, Hemanth ; Chakraborty, Shayok ; Panchanathan, Sethuraman. / Nonlinear embedding transform for unsupervised domain adaptation. Computer Vision – ECCV 2016 Workshops, Proceedings. Vol. 9915 LNCS Springer Verlag, 2016. pp. 451-457 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{6291902d0cd94e30b968249951344c7e,
title = "Nonlinear embedding transform for unsupervised domain adaptation",
abstract = "The problem of domain adaptation (DA) deals with adapting classifier models trained on one data distribution to different data distributions. In this paper, we introduce the Nonlinear Embedding Transform (NET) for unsupervised DA by combining domain alignment along with similarity-based embedding. We also introduce a validation procedure to estimate the model parameters for the NET algorithm using the source data. Comprehensive evaluations on multiple vision datasets demonstrate that the NET algorithm outperforms existing competitive procedures for unsupervised DA.",
keywords = "MMD, Nonlinear embedding, Unsupervised, Validation",
author = "{Demakethepalli Venkateswara}, Hemanth and Shayok Chakraborty and Sethuraman Panchanathan",
year = "2016",
doi = "10.1007/978-3-319-49409-8_36",
language = "English (US)",
isbn = "9783319494081",
volume = "9915 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "451--457",
booktitle = "Computer Vision – ECCV 2016 Workshops, Proceedings",
address = "Germany",

}

TY - GEN

T1 - Nonlinear embedding transform for unsupervised domain adaptation

AU - Demakethepalli Venkateswara, Hemanth

AU - Chakraborty, Shayok

AU - Panchanathan, Sethuraman

PY - 2016

Y1 - 2016

N2 - The problem of domain adaptation (DA) deals with adapting classifier models trained on one data distribution to different data distributions. In this paper, we introduce the Nonlinear Embedding Transform (NET) for unsupervised DA by combining domain alignment along with similarity-based embedding. We also introduce a validation procedure to estimate the model parameters for the NET algorithm using the source data. Comprehensive evaluations on multiple vision datasets demonstrate that the NET algorithm outperforms existing competitive procedures for unsupervised DA.

AB - The problem of domain adaptation (DA) deals with adapting classifier models trained on one data distribution to different data distributions. In this paper, we introduce the Nonlinear Embedding Transform (NET) for unsupervised DA by combining domain alignment along with similarity-based embedding. We also introduce a validation procedure to estimate the model parameters for the NET algorithm using the source data. Comprehensive evaluations on multiple vision datasets demonstrate that the NET algorithm outperforms existing competitive procedures for unsupervised DA.

KW - MMD

KW - Nonlinear embedding

KW - Unsupervised

KW - Validation

UR - http://www.scopus.com/inward/record.url?scp=85005951015&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85005951015&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-49409-8_36

DO - 10.1007/978-3-319-49409-8_36

M3 - Conference contribution

AN - SCOPUS:85005951015

SN - 9783319494081

VL - 9915 LNCS

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

SP - 451

EP - 457

BT - Computer Vision – ECCV 2016 Workshops, Proceedings

PB - Springer Verlag

ER -