Unsupervised source selection for domain adaptation

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Autoren

Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Seiten (von - bis)249-261
Seitenumfang13
FachzeitschriftPhotogrammetric Engineering and Remote Sensing
Jahrgang84
Ausgabenummer5
PublikationsstatusVeröffentlicht - Mai 2018

Abstract

The creation of training sets for supervised machine learning often incurs unsustainable manual costs. Transfer learning (TL) techniques have been proposed as a way to solve this issue by adapting training data from different, but related (source) datasets to the test (target) dataset. A problem in TL is how to quantify the relatedness of a source quickly and robustly. In this work, we present a fast domain similarity measure that captures the relatedness between datasets purely based on unlabeled data. Our method transfers knowledge from multiple sources by generating a weighted combination of domains. We show for multiple datasets that learning on such sources achieves an average overall accuracy closer than 2.5 percent to the results of the target classifier for semantic segmentation tasks. We further apply our method to the task of choosing informative patches from unlabeled datasets. Only labeling these patches enables a reduction in manual work of up to 85 percent.

ASJC Scopus Sachgebiete

Zitieren

Unsupervised source selection for domain adaptation. / Vogt, Karsten; Paul, Andreas; Ostermann, Jörn et al.
in: Photogrammetric Engineering and Remote Sensing, Jahrgang 84, Nr. 5, 05.2018, S. 249-261.

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Vogt K, Paul A, Ostermann J, Rottensteiner F, Heipke C. Unsupervised source selection for domain adaptation. Photogrammetric Engineering and Remote Sensing. 2018 Mai;84(5):249-261. doi: 10.14358/PERS.84.5.249
Vogt, Karsten ; Paul, Andreas ; Ostermann, Jörn et al. / Unsupervised source selection for domain adaptation. in: Photogrammetric Engineering and Remote Sensing. 2018 ; Jahrgang 84, Nr. 5. S. 249-261.
Download
@article{1b7cd1562a6343c8a8d372aac8a1a5c1,
title = "Unsupervised source selection for domain adaptation",
abstract = "The creation of training sets for supervised machine learning often incurs unsustainable manual costs. Transfer learning (TL) techniques have been proposed as a way to solve this issue by adapting training data from different, but related (source) datasets to the test (target) dataset. A problem in TL is how to quantify the relatedness of a source quickly and robustly. In this work, we present a fast domain similarity measure that captures the relatedness between datasets purely based on unlabeled data. Our method transfers knowledge from multiple sources by generating a weighted combination of domains. We show for multiple datasets that learning on such sources achieves an average overall accuracy closer than 2.5 percent to the results of the target classifier for semantic segmentation tasks. We further apply our method to the task of choosing informative patches from unlabeled datasets. Only labeling these patches enables a reduction in manual work of up to 85 percent.",
author = "Karsten Vogt and Andreas Paul and J{\"o}rn Ostermann and Franz Rottensteiner and Christian Heipke",
note = "{\textcopyright} 2018 American Society for Photogrammetry and Remote Sensing",
year = "2018",
month = may,
doi = "10.14358/PERS.84.5.249",
language = "English",
volume = "84",
pages = "249--261",
journal = "Photogrammetric Engineering and Remote Sensing",
issn = "0099-1112",
publisher = "American Society for Photogrammetry and Remote Sensing",
number = "5",

}

Download

TY - JOUR

T1 - Unsupervised source selection for domain adaptation

AU - Vogt, Karsten

AU - Paul, Andreas

AU - Ostermann, Jörn

AU - Rottensteiner, Franz

AU - Heipke, Christian

N1 - © 2018 American Society for Photogrammetry and Remote Sensing

PY - 2018/5

Y1 - 2018/5

N2 - The creation of training sets for supervised machine learning often incurs unsustainable manual costs. Transfer learning (TL) techniques have been proposed as a way to solve this issue by adapting training data from different, but related (source) datasets to the test (target) dataset. A problem in TL is how to quantify the relatedness of a source quickly and robustly. In this work, we present a fast domain similarity measure that captures the relatedness between datasets purely based on unlabeled data. Our method transfers knowledge from multiple sources by generating a weighted combination of domains. We show for multiple datasets that learning on such sources achieves an average overall accuracy closer than 2.5 percent to the results of the target classifier for semantic segmentation tasks. We further apply our method to the task of choosing informative patches from unlabeled datasets. Only labeling these patches enables a reduction in manual work of up to 85 percent.

AB - The creation of training sets for supervised machine learning often incurs unsustainable manual costs. Transfer learning (TL) techniques have been proposed as a way to solve this issue by adapting training data from different, but related (source) datasets to the test (target) dataset. A problem in TL is how to quantify the relatedness of a source quickly and robustly. In this work, we present a fast domain similarity measure that captures the relatedness between datasets purely based on unlabeled data. Our method transfers knowledge from multiple sources by generating a weighted combination of domains. We show for multiple datasets that learning on such sources achieves an average overall accuracy closer than 2.5 percent to the results of the target classifier for semantic segmentation tasks. We further apply our method to the task of choosing informative patches from unlabeled datasets. Only labeling these patches enables a reduction in manual work of up to 85 percent.

UR - http://www.scopus.com/inward/record.url?scp=85047387296&partnerID=8YFLogxK

U2 - 10.14358/PERS.84.5.249

DO - 10.14358/PERS.84.5.249

M3 - Article

AN - SCOPUS:85047387296

VL - 84

SP - 249

EP - 261

JO - Photogrammetric Engineering and Remote Sensing

JF - Photogrammetric Engineering and Remote Sensing

SN - 0099-1112

IS - 5

ER -

Von denselben Autoren