Uncertainty-guided Open-Set Source-Free Unsupervised Domain Adaptation with Target-private Class Segregation

Standard Unsupervised Domain Adaptation (UDA) aims to transfer knowledge from a labeled source domain to an unlabeled target but usually requires simultaneous access to both source and target data. Moreover, UDA approaches commonly assume that source and target domains share the same labels space. Y...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2024-04
Hauptverfasser: Litrico, Mattia, Talon, Davide, Battiato, Sebastiano, Alessio Del Bue, Mario Valerio Giuffrida, Morerio, Pietro
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Litrico, Mattia
Talon, Davide
Battiato, Sebastiano
Alessio Del Bue
Mario Valerio Giuffrida
Morerio, Pietro
description Standard Unsupervised Domain Adaptation (UDA) aims to transfer knowledge from a labeled source domain to an unlabeled target but usually requires simultaneous access to both source and target data. Moreover, UDA approaches commonly assume that source and target domains share the same labels space. Yet, these two assumptions are hardly satisfied in real-world scenarios. This paper considers the more challenging Source-Free Open-set Domain Adaptation (SF-OSDA) setting, where both assumptions are dropped. We propose a novel approach for SF-OSDA that exploits the granularity of target-private categories by segregating their samples into multiple unknown classes. Starting from an initial clustering-based assignment, our method progressively improves the segregation of target-private samples by refining their pseudo-labels with the guide of an uncertainty-based sample selection module. Additionally, we propose a novel contrastive loss, named NL-InfoNCELoss, that, integrating negative learning into self-supervised contrastive learning, enhances the model robustness to noisy pseudo-labels. Extensive experiments on benchmark datasets demonstrate the superiority of the proposed method over existing approaches, establishing new state-of-the-art performance. Notably, additional analyses show that our method is able to learn the underlying semantics of novel classes, opening the possibility to perform novel class discovery.
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_3040136379</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3040136379</sourcerecordid><originalsourceid>FETCH-proquest_journals_30401363793</originalsourceid><addsrcrecordid>eNqNjb0KwjAURoMgWNR3CDgH0qT-jVIVN4fWuQR7jZGaxORW8e0N4gM4fcM5h29AMiFlzlaFECMyjfHGOReLpZjPZUa6kz1DQGUsvpnuTQstPXqwrAKklevDGdg-ANCTjb2H8DQxGVt3TwXdtMqjQuMsfRm80loFDch8ME-FQMtOxUgr0AH015qQ4UV1Eaa_HZPZfleXh1S4Rw8Rm1t6tAk1khc8lwu5XMv_rA_QOEm8</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3040136379</pqid></control><display><type>article</type><title>Uncertainty-guided Open-Set Source-Free Unsupervised Domain Adaptation with Target-private Class Segregation</title><source>Free E- Journals</source><creator>Litrico, Mattia ; Talon, Davide ; Battiato, Sebastiano ; Alessio Del Bue ; Mario Valerio Giuffrida ; Morerio, Pietro</creator><creatorcontrib>Litrico, Mattia ; Talon, Davide ; Battiato, Sebastiano ; Alessio Del Bue ; Mario Valerio Giuffrida ; Morerio, Pietro</creatorcontrib><description>Standard Unsupervised Domain Adaptation (UDA) aims to transfer knowledge from a labeled source domain to an unlabeled target but usually requires simultaneous access to both source and target data. Moreover, UDA approaches commonly assume that source and target domains share the same labels space. Yet, these two assumptions are hardly satisfied in real-world scenarios. This paper considers the more challenging Source-Free Open-set Domain Adaptation (SF-OSDA) setting, where both assumptions are dropped. We propose a novel approach for SF-OSDA that exploits the granularity of target-private categories by segregating their samples into multiple unknown classes. Starting from an initial clustering-based assignment, our method progressively improves the segregation of target-private samples by refining their pseudo-labels with the guide of an uncertainty-based sample selection module. Additionally, we propose a novel contrastive loss, named NL-InfoNCELoss, that, integrating negative learning into self-supervised contrastive learning, enhances the model robustness to noisy pseudo-labels. Extensive experiments on benchmark datasets demonstrate the superiority of the proposed method over existing approaches, establishing new state-of-the-art performance. Notably, additional analyses show that our method is able to learn the underlying semantics of novel classes, opening the possibility to perform novel class discovery.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Adaptation ; Clustering ; Knowledge management ; Labels ; Learning ; Semantics ; Uncertainty</subject><ispartof>arXiv.org, 2024-04</ispartof><rights>2024. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>776,780</link.rule.ids></links><search><creatorcontrib>Litrico, Mattia</creatorcontrib><creatorcontrib>Talon, Davide</creatorcontrib><creatorcontrib>Battiato, Sebastiano</creatorcontrib><creatorcontrib>Alessio Del Bue</creatorcontrib><creatorcontrib>Mario Valerio Giuffrida</creatorcontrib><creatorcontrib>Morerio, Pietro</creatorcontrib><title>Uncertainty-guided Open-Set Source-Free Unsupervised Domain Adaptation with Target-private Class Segregation</title><title>arXiv.org</title><description>Standard Unsupervised Domain Adaptation (UDA) aims to transfer knowledge from a labeled source domain to an unlabeled target but usually requires simultaneous access to both source and target data. Moreover, UDA approaches commonly assume that source and target domains share the same labels space. Yet, these two assumptions are hardly satisfied in real-world scenarios. This paper considers the more challenging Source-Free Open-set Domain Adaptation (SF-OSDA) setting, where both assumptions are dropped. We propose a novel approach for SF-OSDA that exploits the granularity of target-private categories by segregating their samples into multiple unknown classes. Starting from an initial clustering-based assignment, our method progressively improves the segregation of target-private samples by refining their pseudo-labels with the guide of an uncertainty-based sample selection module. Additionally, we propose a novel contrastive loss, named NL-InfoNCELoss, that, integrating negative learning into self-supervised contrastive learning, enhances the model robustness to noisy pseudo-labels. Extensive experiments on benchmark datasets demonstrate the superiority of the proposed method over existing approaches, establishing new state-of-the-art performance. Notably, additional analyses show that our method is able to learn the underlying semantics of novel classes, opening the possibility to perform novel class discovery.</description><subject>Adaptation</subject><subject>Clustering</subject><subject>Knowledge management</subject><subject>Labels</subject><subject>Learning</subject><subject>Semantics</subject><subject>Uncertainty</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>BENPR</sourceid><recordid>eNqNjb0KwjAURoMgWNR3CDgH0qT-jVIVN4fWuQR7jZGaxORW8e0N4gM4fcM5h29AMiFlzlaFECMyjfHGOReLpZjPZUa6kz1DQGUsvpnuTQstPXqwrAKklevDGdg-ANCTjb2H8DQxGVt3TwXdtMqjQuMsfRm80loFDch8ME-FQMtOxUgr0AH015qQ4UV1Eaa_HZPZfleXh1S4Rw8Rm1t6tAk1khc8lwu5XMv_rA_QOEm8</recordid><startdate>20240416</startdate><enddate>20240416</enddate><creator>Litrico, Mattia</creator><creator>Talon, Davide</creator><creator>Battiato, Sebastiano</creator><creator>Alessio Del Bue</creator><creator>Mario Valerio Giuffrida</creator><creator>Morerio, Pietro</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20240416</creationdate><title>Uncertainty-guided Open-Set Source-Free Unsupervised Domain Adaptation with Target-private Class Segregation</title><author>Litrico, Mattia ; Talon, Davide ; Battiato, Sebastiano ; Alessio Del Bue ; Mario Valerio Giuffrida ; Morerio, Pietro</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_30401363793</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Adaptation</topic><topic>Clustering</topic><topic>Knowledge management</topic><topic>Labels</topic><topic>Learning</topic><topic>Semantics</topic><topic>Uncertainty</topic><toplevel>online_resources</toplevel><creatorcontrib>Litrico, Mattia</creatorcontrib><creatorcontrib>Talon, Davide</creatorcontrib><creatorcontrib>Battiato, Sebastiano</creatorcontrib><creatorcontrib>Alessio Del Bue</creatorcontrib><creatorcontrib>Mario Valerio Giuffrida</creatorcontrib><creatorcontrib>Morerio, Pietro</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Litrico, Mattia</au><au>Talon, Davide</au><au>Battiato, Sebastiano</au><au>Alessio Del Bue</au><au>Mario Valerio Giuffrida</au><au>Morerio, Pietro</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Uncertainty-guided Open-Set Source-Free Unsupervised Domain Adaptation with Target-private Class Segregation</atitle><jtitle>arXiv.org</jtitle><date>2024-04-16</date><risdate>2024</risdate><eissn>2331-8422</eissn><abstract>Standard Unsupervised Domain Adaptation (UDA) aims to transfer knowledge from a labeled source domain to an unlabeled target but usually requires simultaneous access to both source and target data. Moreover, UDA approaches commonly assume that source and target domains share the same labels space. Yet, these two assumptions are hardly satisfied in real-world scenarios. This paper considers the more challenging Source-Free Open-set Domain Adaptation (SF-OSDA) setting, where both assumptions are dropped. We propose a novel approach for SF-OSDA that exploits the granularity of target-private categories by segregating their samples into multiple unknown classes. Starting from an initial clustering-based assignment, our method progressively improves the segregation of target-private samples by refining their pseudo-labels with the guide of an uncertainty-based sample selection module. Additionally, we propose a novel contrastive loss, named NL-InfoNCELoss, that, integrating negative learning into self-supervised contrastive learning, enhances the model robustness to noisy pseudo-labels. Extensive experiments on benchmark datasets demonstrate the superiority of the proposed method over existing approaches, establishing new state-of-the-art performance. Notably, additional analyses show that our method is able to learn the underlying semantics of novel classes, opening the possibility to perform novel class discovery.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2024-04
issn 2331-8422
language eng
recordid cdi_proquest_journals_3040136379
source Free E- Journals
subjects Adaptation
Clustering
Knowledge management
Labels
Learning
Semantics
Uncertainty
title Uncertainty-guided Open-Set Source-Free Unsupervised Domain Adaptation with Target-private Class Segregation
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-05T18%3A30%3A38IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Uncertainty-guided%20Open-Set%20Source-Free%20Unsupervised%20Domain%20Adaptation%20with%20Target-private%20Class%20Segregation&rft.jtitle=arXiv.org&rft.au=Litrico,%20Mattia&rft.date=2024-04-16&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E3040136379%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=3040136379&rft_id=info:pmid/&rfr_iscdi=true