Nonlinear embedding transform for unsupervised domain adaptation

Hemanth Demakethepalli Venkateswara; Shayok Chakraborty; Sethuraman Panchanathan

doi:10.1007/978-3-319-49409-8_36

Nonlinear embedding transform for unsupervised domain adaptation

Hemanth Demakethepalli Venkateswara, Shayok Chakraborty, Sethuraman Panchanathan

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Scopus citations

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 language	English (US)
Title of host publication	Computer Vision – ECCV 2016 Workshops, Proceedings
Editors	Gang Hua, Herve Jegou
Publisher	Springer Verlag
Pages	451-457
Number of pages	7
ISBN (Print)	9783319494081
DOIs	https://doi.org/10.1007/978-3-319-49409-8_36
State	Published - 2016
Event	Computer Vision - ECCV 2016 Workshops, Proceedings - Duration: Jan 1 2016 → …

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9915 LNCS

Conference

Conference	Computer Vision - ECCV 2016 Workshops, Proceedings
Period	1/1/16 → …

Keywords

MMD
Nonlinear embedding
Unsupervised
Validation

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

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

Cite this

Demakethepalli Venkateswara, H., Chakraborty, S., & Panchanathan, S. (2016). Nonlinear embedding transform for unsupervised domain adaptation. In G. Hua, & H. Jegou (Eds.), Computer Vision – ECCV 2016 Workshops, Proceedings (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. ed. / Gang Hua; Herve Jegou. 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 proceeding › Conference contribution

Demakethepalli Venkateswara, H, Chakraborty, S & Panchanathan, S 2016, Nonlinear embedding transform for unsupervised domain adaptation. in G Hua & H Jegou (eds), Computer Vision – ECCV 2016 Workshops, Proceedings. 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, Computer Vision - ECCV 2016 Workshops, Proceedings, 1/1/16. 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 Hua G, Jegou H, editors, Computer Vision – ECCV 2016 Workshops, Proceedings. Springer Verlag. 2016. p. 451-457. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 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. editor / Gang Hua ; Herve Jegou. Springer Verlag, 2016. pp. 451-457 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{31170783399149978d9e1f99cbb2a0ba,

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",

note = "Funding Information: We have proposed the NET algorithm for unsupervised DA along with a procedure for generating a validation set for model selection using the source data. Both the validation procedure and NET have better recognition accuracies than competitive visual DA methods across multiple vision based datasets. This material is based upon work supported by the National Science Foundation (NSF) under Grant No:1116360. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF. Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2016.; Computer Vision - ECCV 2016 Workshops, Proceedings ; Conference date: 01-01-2016",

year = "2016",

doi = "10.1007/978-3-319-49409-8_36",

language = "English (US)",

isbn = "9783319494081",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "451--457",

editor = "Gang Hua and Herve Jegou",

booktitle = "Computer Vision – ECCV 2016 Workshops, Proceedings",

}

TY - GEN

T1 - Nonlinear embedding transform for unsupervised domain adaptation

AU - Demakethepalli Venkateswara, Hemanth

AU - Chakraborty, Shayok

AU - Panchanathan, Sethuraman

N1 - Funding Information: We have proposed the NET algorithm for unsupervised DA along with a procedure for generating a validation set for model selection using the source data. Both the validation procedure and NET have better recognition accuracies than competitive visual DA methods across multiple vision based datasets. This material is based upon work supported by the National Science Foundation (NSF) under Grant No:1116360. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF. Publisher Copyright: © Springer International Publishing Switzerland 2016.

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

SN - 9783319494081

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

A2 - Hua, Gang

A2 - Jegou, Herve

PB - Springer Verlag

T2 - Computer Vision - ECCV 2016 Workshops, Proceedings

Y2 - 1 January 2016

ER -

Nonlinear embedding transform for unsupervised domain adaptation

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this