Multiple Subspace Alignment Improves Domain Adaptation

Kowshik Thopalli, Rushil Anirudh, Jayaraman J. Thiagarajan, Pavan Turaga

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

Abstract

We present a novel unsupervised domain adaptation (DA) method for cross-domain visual recognition. Though subspace methods have found success in DA, their performance is often limited due to the assumption of approximating an entire dataset using a single low-dimensional subspace. Instead, we develop a method to effectively represent the source and target datasets via a collection of low-dimensional subspaces, and subsequently align them by exploiting the natural geometry of the space of subspaces, on the Grassmann manifold. We demonstrate the effectiveness of this approach, using empirical studies on two widely used benchmarks,with performance on par or better than the performance of the state of the art domain adaptation methods.

Original languageEnglish (US)
Title of host publication2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages3552-3556
Number of pages5
ISBN (Electronic)9781479981311
DOIs
StatePublished - May 1 2019
Event44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Brighton, United Kingdom
Duration: May 12 2019May 17 2019

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume2019-May
ISSN (Print)1520-6149

Conference

Conference44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019
CountryUnited Kingdom
CityBrighton
Period5/12/195/17/19

Fingerprint

Geometry

Keywords

  • Domain Adaptation
  • Grassmann manifold
  • Unsupervised Learning

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Thopalli, K., Anirudh, R., Thiagarajan, J. J., & Turaga, P. (2019). Multiple Subspace Alignment Improves Domain Adaptation. In 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings (pp. 3552-3556). [8682420] (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2019-May). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2019.8682420

Multiple Subspace Alignment Improves Domain Adaptation. / Thopalli, Kowshik; Anirudh, Rushil; Thiagarajan, Jayaraman J.; Turaga, Pavan.

2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. p. 3552-3556 8682420 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2019-May).

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

Thopalli, K, Anirudh, R, Thiagarajan, JJ & Turaga, P 2019, Multiple Subspace Alignment Improves Domain Adaptation. in 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings., 8682420, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2019-May, Institute of Electrical and Electronics Engineers Inc., pp. 3552-3556, 44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019, Brighton, United Kingdom, 5/12/19. https://doi.org/10.1109/ICASSP.2019.8682420
Thopalli K, Anirudh R, Thiagarajan JJ, Turaga P. Multiple Subspace Alignment Improves Domain Adaptation. In 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. p. 3552-3556. 8682420. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2019.8682420
Thopalli, Kowshik ; Anirudh, Rushil ; Thiagarajan, Jayaraman J. ; Turaga, Pavan. / Multiple Subspace Alignment Improves Domain Adaptation. 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 3552-3556 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).
@inproceedings{e7a5d5ab54e3433e9d1526a5cc2ec4be,
title = "Multiple Subspace Alignment Improves Domain Adaptation",
abstract = "We present a novel unsupervised domain adaptation (DA) method for cross-domain visual recognition. Though subspace methods have found success in DA, their performance is often limited due to the assumption of approximating an entire dataset using a single low-dimensional subspace. Instead, we develop a method to effectively represent the source and target datasets via a collection of low-dimensional subspaces, and subsequently align them by exploiting the natural geometry of the space of subspaces, on the Grassmann manifold. We demonstrate the effectiveness of this approach, using empirical studies on two widely used benchmarks,with performance on par or better than the performance of the state of the art domain adaptation methods.",
keywords = "Domain Adaptation, Grassmann manifold, Unsupervised Learning",
author = "Kowshik Thopalli and Rushil Anirudh and Thiagarajan, {Jayaraman J.} and Pavan Turaga",
year = "2019",
month = "5",
day = "1",
doi = "10.1109/ICASSP.2019.8682420",
language = "English (US)",
series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "3552--3556",
booktitle = "2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings",

}

TY - GEN

T1 - Multiple Subspace Alignment Improves Domain Adaptation

AU - Thopalli, Kowshik

AU - Anirudh, Rushil

AU - Thiagarajan, Jayaraman J.

AU - Turaga, Pavan

PY - 2019/5/1

Y1 - 2019/5/1

N2 - We present a novel unsupervised domain adaptation (DA) method for cross-domain visual recognition. Though subspace methods have found success in DA, their performance is often limited due to the assumption of approximating an entire dataset using a single low-dimensional subspace. Instead, we develop a method to effectively represent the source and target datasets via a collection of low-dimensional subspaces, and subsequently align them by exploiting the natural geometry of the space of subspaces, on the Grassmann manifold. We demonstrate the effectiveness of this approach, using empirical studies on two widely used benchmarks,with performance on par or better than the performance of the state of the art domain adaptation methods.

AB - We present a novel unsupervised domain adaptation (DA) method for cross-domain visual recognition. Though subspace methods have found success in DA, their performance is often limited due to the assumption of approximating an entire dataset using a single low-dimensional subspace. Instead, we develop a method to effectively represent the source and target datasets via a collection of low-dimensional subspaces, and subsequently align them by exploiting the natural geometry of the space of subspaces, on the Grassmann manifold. We demonstrate the effectiveness of this approach, using empirical studies on two widely used benchmarks,with performance on par or better than the performance of the state of the art domain adaptation methods.

KW - Domain Adaptation

KW - Grassmann manifold

KW - Unsupervised Learning

UR - http://www.scopus.com/inward/record.url?scp=85068958606&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068958606&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2019.8682420

DO - 10.1109/ICASSP.2019.8682420

M3 - Conference contribution

AN - SCOPUS:85068958606

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 3552

EP - 3556

BT - 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -