Key Design Choices in Source-Free Unsupervised Domain Adaptation: An In-depth Empirical Analysis

This study provides a comprehensive benchmark framework for Source-Free Unsupervised Domain Adaptation (SF-UDA) in image classification, aiming to achieve a rigorous empirical understanding of the complex relationships between multiple key design factors in SF-UDA methods. The study empirically exam...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Maracani, Andrea, Camoriano, Raffaello, Maiettini, Elisa, Talon, Davide, Rosasco, Lorenzo, Natale, Lorenzo
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Maracani, Andrea
Camoriano, Raffaello
Maiettini, Elisa
Talon, Davide
Rosasco, Lorenzo
Natale, Lorenzo
description This study provides a comprehensive benchmark framework for Source-Free Unsupervised Domain Adaptation (SF-UDA) in image classification, aiming to achieve a rigorous empirical understanding of the complex relationships between multiple key design factors in SF-UDA methods. The study empirically examines a diverse set of SF-UDA techniques, assessing their consistency across datasets, sensitivity to specific hyperparameters, and applicability across different families of backbone architectures. Moreover, it exhaustively evaluates pre-training datasets and strategies, particularly focusing on both supervised and self-supervised methods, as well as the impact of fine-tuning on the source domain. Our analysis also highlights gaps in existing benchmark practices, guiding SF-UDA research towards more effective and general approaches. It emphasizes the importance of backbone architecture and pre-training dataset selection on SF-UDA performance, serving as an essential reference and providing key insights. Lastly, we release the source code of our experimental framework. This facilitates the construction, training, and testing of SF-UDA methods, enabling systematic large-scale experimental analysis and supporting further research efforts in this field.
doi_str_mv 10.48550/arxiv.2402.16090
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2402_16090</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2402_16090</sourcerecordid><originalsourceid>FETCH-LOGICAL-a670-7ed8b8241e0b4f2c13b58389ec98ed31788f5476cd568cc99f26ca35c49229303</originalsourceid><addsrcrecordid>eNotz71OwzAcBHAvDKjwAEz4BRIcfyQ2W5S2ULUSA-2cOvY_1FLiRHZakbenFKaT7qSTfgg9ZSTlUgjyosO3u6SUE5pmOVHkHh23MOMlRPflcXUanIGIncefwzkYSNYBAB98PI8QLi6Cxcuh19e9tHqc9OQG_4pLjzc-sTBOJ7zqRxec0d211d0cXXxAd63uIjz-5wLt16t99Z7sPt42VblLdF6QpAArG0l5BqThLTUZa4RkUoFREizLCilbwYvcWJFLY5RqaW40E4YrShUjbIGe_25vxHoMrtdhrn-p9Y3KfgAqE04q</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Key Design Choices in Source-Free Unsupervised Domain Adaptation: An In-depth Empirical Analysis</title><source>arXiv.org</source><creator>Maracani, Andrea ; Camoriano, Raffaello ; Maiettini, Elisa ; Talon, Davide ; Rosasco, Lorenzo ; Natale, Lorenzo</creator><creatorcontrib>Maracani, Andrea ; Camoriano, Raffaello ; Maiettini, Elisa ; Talon, Davide ; Rosasco, Lorenzo ; Natale, Lorenzo</creatorcontrib><description>This study provides a comprehensive benchmark framework for Source-Free Unsupervised Domain Adaptation (SF-UDA) in image classification, aiming to achieve a rigorous empirical understanding of the complex relationships between multiple key design factors in SF-UDA methods. The study empirically examines a diverse set of SF-UDA techniques, assessing their consistency across datasets, sensitivity to specific hyperparameters, and applicability across different families of backbone architectures. Moreover, it exhaustively evaluates pre-training datasets and strategies, particularly focusing on both supervised and self-supervised methods, as well as the impact of fine-tuning on the source domain. Our analysis also highlights gaps in existing benchmark practices, guiding SF-UDA research towards more effective and general approaches. It emphasizes the importance of backbone architecture and pre-training dataset selection on SF-UDA performance, serving as an essential reference and providing key insights. Lastly, we release the source code of our experimental framework. This facilitates the construction, training, and testing of SF-UDA methods, enabling systematic large-scale experimental analysis and supporting further research efforts in this field.</description><identifier>DOI: 10.48550/arxiv.2402.16090</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition ; Computer Science - Learning</subject><creationdate>2024-02</creationdate><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2402.16090$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2402.16090$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Maracani, Andrea</creatorcontrib><creatorcontrib>Camoriano, Raffaello</creatorcontrib><creatorcontrib>Maiettini, Elisa</creatorcontrib><creatorcontrib>Talon, Davide</creatorcontrib><creatorcontrib>Rosasco, Lorenzo</creatorcontrib><creatorcontrib>Natale, Lorenzo</creatorcontrib><title>Key Design Choices in Source-Free Unsupervised Domain Adaptation: An In-depth Empirical Analysis</title><description>This study provides a comprehensive benchmark framework for Source-Free Unsupervised Domain Adaptation (SF-UDA) in image classification, aiming to achieve a rigorous empirical understanding of the complex relationships between multiple key design factors in SF-UDA methods. The study empirically examines a diverse set of SF-UDA techniques, assessing their consistency across datasets, sensitivity to specific hyperparameters, and applicability across different families of backbone architectures. Moreover, it exhaustively evaluates pre-training datasets and strategies, particularly focusing on both supervised and self-supervised methods, as well as the impact of fine-tuning on the source domain. Our analysis also highlights gaps in existing benchmark practices, guiding SF-UDA research towards more effective and general approaches. It emphasizes the importance of backbone architecture and pre-training dataset selection on SF-UDA performance, serving as an essential reference and providing key insights. Lastly, we release the source code of our experimental framework. This facilitates the construction, training, and testing of SF-UDA methods, enabling systematic large-scale experimental analysis and supporting further research efforts in this field.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotz71OwzAcBHAvDKjwAEz4BRIcfyQ2W5S2ULUSA-2cOvY_1FLiRHZakbenFKaT7qSTfgg9ZSTlUgjyosO3u6SUE5pmOVHkHh23MOMlRPflcXUanIGIncefwzkYSNYBAB98PI8QLi6Cxcuh19e9tHqc9OQG_4pLjzc-sTBOJ7zqRxec0d211d0cXXxAd63uIjz-5wLt16t99Z7sPt42VblLdF6QpAArG0l5BqThLTUZa4RkUoFREizLCilbwYvcWJFLY5RqaW40E4YrShUjbIGe_25vxHoMrtdhrn-p9Y3KfgAqE04q</recordid><startdate>20240225</startdate><enddate>20240225</enddate><creator>Maracani, Andrea</creator><creator>Camoriano, Raffaello</creator><creator>Maiettini, Elisa</creator><creator>Talon, Davide</creator><creator>Rosasco, Lorenzo</creator><creator>Natale, Lorenzo</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20240225</creationdate><title>Key Design Choices in Source-Free Unsupervised Domain Adaptation: An In-depth Empirical Analysis</title><author>Maracani, Andrea ; Camoriano, Raffaello ; Maiettini, Elisa ; Talon, Davide ; Rosasco, Lorenzo ; Natale, Lorenzo</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a670-7ed8b8241e0b4f2c13b58389ec98ed31788f5476cd568cc99f26ca35c49229303</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Maracani, Andrea</creatorcontrib><creatorcontrib>Camoriano, Raffaello</creatorcontrib><creatorcontrib>Maiettini, Elisa</creatorcontrib><creatorcontrib>Talon, Davide</creatorcontrib><creatorcontrib>Rosasco, Lorenzo</creatorcontrib><creatorcontrib>Natale, Lorenzo</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Maracani, Andrea</au><au>Camoriano, Raffaello</au><au>Maiettini, Elisa</au><au>Talon, Davide</au><au>Rosasco, Lorenzo</au><au>Natale, Lorenzo</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Key Design Choices in Source-Free Unsupervised Domain Adaptation: An In-depth Empirical Analysis</atitle><date>2024-02-25</date><risdate>2024</risdate><abstract>This study provides a comprehensive benchmark framework for Source-Free Unsupervised Domain Adaptation (SF-UDA) in image classification, aiming to achieve a rigorous empirical understanding of the complex relationships between multiple key design factors in SF-UDA methods. The study empirically examines a diverse set of SF-UDA techniques, assessing their consistency across datasets, sensitivity to specific hyperparameters, and applicability across different families of backbone architectures. Moreover, it exhaustively evaluates pre-training datasets and strategies, particularly focusing on both supervised and self-supervised methods, as well as the impact of fine-tuning on the source domain. Our analysis also highlights gaps in existing benchmark practices, guiding SF-UDA research towards more effective and general approaches. It emphasizes the importance of backbone architecture and pre-training dataset selection on SF-UDA performance, serving as an essential reference and providing key insights. Lastly, we release the source code of our experimental framework. This facilitates the construction, training, and testing of SF-UDA methods, enabling systematic large-scale experimental analysis and supporting further research efforts in this field.</abstract><doi>10.48550/arxiv.2402.16090</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2402.16090
ispartof
issn
language eng
recordid cdi_arxiv_primary_2402_16090
source arXiv.org
subjects Computer Science - Computer Vision and Pattern Recognition
Computer Science - Learning
title Key Design Choices in Source-Free Unsupervised Domain Adaptation: An In-depth Empirical Analysis
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-27T17%3A28%3A03IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Key%20Design%20Choices%20in%20Source-Free%20Unsupervised%20Domain%20Adaptation:%20An%20In-depth%20Empirical%20Analysis&rft.au=Maracani,%20Andrea&rft.date=2024-02-25&rft_id=info:doi/10.48550/arxiv.2402.16090&rft_dat=%3Carxiv_GOX%3E2402_16090%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true