Open-set domain adaptation by deconfounding domain gaps

Open-Set Domain Adaptation (OSDA) aims to adapt the model trained on a source domain to the recognition tasks in a target domain while shielding any distractions caused by open-set classes, i.e., the classes “unknown” to the source model. Compared to standard DA, the key of OSDA lies in the separati...

Full description

Saved in:
Bibliographic Details
Main Authors: ZHAO, Xin, WANG, Shengsheng, SUN, Qianru
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2023
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/7556
https://ink.library.smu.edu.sg/context/sis_research/article/8559/viewcontent/Open_Set_Domain_Adaptation_by_Deconfounding_Domain_Gaps__NeuroComputing_.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.sis_research-8559
record_format dspace
spelling sg-smu-ink.sis_research-85592024-02-28T05:34:31Z Open-set domain adaptation by deconfounding domain gaps ZHAO, Xin WANG, Shengsheng SUN, Qianru Open-Set Domain Adaptation (OSDA) aims to adapt the model trained on a source domain to the recognition tasks in a target domain while shielding any distractions caused by open-set classes, i.e., the classes “unknown” to the source model. Compared to standard DA, the key of OSDA lies in the separation between known and unknown classes. Existing OSDA methods often fail the separation because of overlooking the confounders (i.e., the domain gaps), which means their recognition of “unknown classes” is not because of class semantics but domain difference (e.g., styles and contexts). We address this issue by explicitly deconfounding domain gaps (DDP) during class separation and domain adaptation in OSDA. The mechanism of DDP is to transfer domain-related styles and contexts from the target domain to the source domain. It enables the model to recognize a class as known (or unknown) because of the class semantics rather than the confusion caused by spurious styles or contexts. In addition, we propose a module of ensembling multiple transformations (EMT) to produce calibrated recognition scores, i.e., reliable normality scores, for the samples in the target domain. Extensive experiments on two standard benchmarks verify that our proposed method outperforms a wide range of OSDA methods, because of its advanced ability of correctly recognizing unknown classes. 2023-04-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/7556 info:doi/10.1007/s10489-022-03805-9 https://ink.library.smu.edu.sg/context/sis_research/article/8559/viewcontent/Open_Set_Domain_Adaptation_by_Deconfounding_Domain_Gaps__NeuroComputing_.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Open-set domain adaptation Image classification Artificial Intelligence and Robotics Databases and Information Systems Graphics and Human Computer Interfaces
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Open-set domain adaptation
Image classification
Artificial Intelligence and Robotics
Databases and Information Systems
Graphics and Human Computer Interfaces
spellingShingle Open-set domain adaptation
Image classification
Artificial Intelligence and Robotics
Databases and Information Systems
Graphics and Human Computer Interfaces
ZHAO, Xin
WANG, Shengsheng
SUN, Qianru
Open-set domain adaptation by deconfounding domain gaps
description Open-Set Domain Adaptation (OSDA) aims to adapt the model trained on a source domain to the recognition tasks in a target domain while shielding any distractions caused by open-set classes, i.e., the classes “unknown” to the source model. Compared to standard DA, the key of OSDA lies in the separation between known and unknown classes. Existing OSDA methods often fail the separation because of overlooking the confounders (i.e., the domain gaps), which means their recognition of “unknown classes” is not because of class semantics but domain difference (e.g., styles and contexts). We address this issue by explicitly deconfounding domain gaps (DDP) during class separation and domain adaptation in OSDA. The mechanism of DDP is to transfer domain-related styles and contexts from the target domain to the source domain. It enables the model to recognize a class as known (or unknown) because of the class semantics rather than the confusion caused by spurious styles or contexts. In addition, we propose a module of ensembling multiple transformations (EMT) to produce calibrated recognition scores, i.e., reliable normality scores, for the samples in the target domain. Extensive experiments on two standard benchmarks verify that our proposed method outperforms a wide range of OSDA methods, because of its advanced ability of correctly recognizing unknown classes.
format text
author ZHAO, Xin
WANG, Shengsheng
SUN, Qianru
author_facet ZHAO, Xin
WANG, Shengsheng
SUN, Qianru
author_sort ZHAO, Xin
title Open-set domain adaptation by deconfounding domain gaps
title_short Open-set domain adaptation by deconfounding domain gaps
title_full Open-set domain adaptation by deconfounding domain gaps
title_fullStr Open-set domain adaptation by deconfounding domain gaps
title_full_unstemmed Open-set domain adaptation by deconfounding domain gaps
title_sort open-set domain adaptation by deconfounding domain gaps
publisher Institutional Knowledge at Singapore Management University
publishDate 2023
url https://ink.library.smu.edu.sg/sis_research/7556
https://ink.library.smu.edu.sg/context/sis_research/article/8559/viewcontent/Open_Set_Domain_Adaptation_by_Deconfounding_Domain_Gaps__NeuroComputing_.pdf
_version_ 1794549717566226432