AIR-Net: Adaptive and Implicit Regularization Neural Network for Matrix Completion

The explicit low-rank regularization, e.g., nuclear norm regularization, has been widely used in imaging sciences. However, it has been found that implicit regularization outperforms explicit ones in various image processing tasks. Another issue is that the fixed explicit regularization limits the a...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2022-06
Hauptverfasser: Li, Zhemin, Sun, Tao, Wang, Hongxia, Wang, Bao
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 Li, Zhemin
Sun, Tao
Wang, Hongxia
Wang, Bao
description The explicit low-rank regularization, e.g., nuclear norm regularization, has been widely used in imaging sciences. However, it has been found that implicit regularization outperforms explicit ones in various image processing tasks. Another issue is that the fixed explicit regularization limits the applicability to broad images since different images favor different features captured by different explicit regularizations. As such, this paper proposes a new adaptive and implicit low-rank regularization that captures the low-rank prior dynamically from the training data. The core of our new adaptive and implicit low-rank regularization is parameterizing the Laplacian matrix in the Dirichlet energy-based regularization, which we call the regularization \textit{AIR}. Theoretically, we show that the adaptive regularization of AIR enhances the implicit regularization and vanishes at the end of training. We validate AIR's effectiveness on various benchmark tasks, indicating that the AIR is particularly favorable for the scenarios when the missing entries are non-uniform. The code can be found at https://github.com/lizhemin15/AIR-Net
doi_str_mv 10.48550/arxiv.2110.07557
format Article
fullrecord <record><control><sourceid>proquest_arxiv</sourceid><recordid>TN_cdi_arxiv_primary_2110_07557</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2582280138</sourcerecordid><originalsourceid>FETCH-LOGICAL-a957-fb041b48b2829766acb378de043a4b52faaab00396d2306f0f61f2669dd89ba73</originalsourceid><addsrcrecordid>eNotj11PwjAUhhsTEwnyA7yyidfD7nT9mHeEqJAgJoT75ZS1pjjY7DpEf70DvHqTN885eR9C7lI2zrQQ7BHD0R_GkPYFU0KoKzIAztNEZwA3ZNS2W8YYSAVC8AFZTearZGnjE52U2ER_sBT3JZ3vmspvfKQr-9FVGPwvRl_v6dJ2Aas-4ncdPqmrA33DGPyRTuv-xJ6gW3LtsGrt6D-HZP3yvJ7OksX763w6WSSYC5U4w7LUZNqAhlxJiRvDlS4tyzhmRoBDRMMYz2UJnEnHnEwdSJmXpc4NKj4k95e3Z-GiCX6H4ac4iRdn8Z54uBBNqL8628ZiW3dh328qQGgAzVKu-R-j7VrM</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2582280138</pqid></control><display><type>article</type><title>AIR-Net: Adaptive and Implicit Regularization Neural Network for Matrix Completion</title><source>arXiv.org</source><source>Free E- Journals</source><creator>Li, Zhemin ; Sun, Tao ; Wang, Hongxia ; Wang, Bao</creator><creatorcontrib>Li, Zhemin ; Sun, Tao ; Wang, Hongxia ; Wang, Bao</creatorcontrib><description>The explicit low-rank regularization, e.g., nuclear norm regularization, has been widely used in imaging sciences. However, it has been found that implicit regularization outperforms explicit ones in various image processing tasks. Another issue is that the fixed explicit regularization limits the applicability to broad images since different images favor different features captured by different explicit regularizations. As such, this paper proposes a new adaptive and implicit low-rank regularization that captures the low-rank prior dynamically from the training data. The core of our new adaptive and implicit low-rank regularization is parameterizing the Laplacian matrix in the Dirichlet energy-based regularization, which we call the regularization \textit{AIR}. Theoretically, we show that the adaptive regularization of AIR enhances the implicit regularization and vanishes at the end of training. We validate AIR's effectiveness on various benchmark tasks, indicating that the AIR is particularly favorable for the scenarios when the missing entries are non-uniform. The code can be found at https://github.com/lizhemin15/AIR-Net</description><identifier>EISSN: 2331-8422</identifier><identifier>DOI: 10.48550/arxiv.2110.07557</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Computer Science - Learning ; Inverse problems ; Matrix representation ; Neural networks ; Regularization ; Statistics - Machine Learning</subject><ispartof>arXiv.org, 2022-06</ispartof><rights>2022. 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><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>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,784,885,27924</link.rule.ids><backlink>$$Uhttps://doi.org/10.1137/22M1489228$$DView published paper (Access to full text may be restricted)$$Hfree_for_read</backlink><backlink>$$Uhttps://doi.org/10.48550/arXiv.2110.07557$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Li, Zhemin</creatorcontrib><creatorcontrib>Sun, Tao</creatorcontrib><creatorcontrib>Wang, Hongxia</creatorcontrib><creatorcontrib>Wang, Bao</creatorcontrib><title>AIR-Net: Adaptive and Implicit Regularization Neural Network for Matrix Completion</title><title>arXiv.org</title><description>The explicit low-rank regularization, e.g., nuclear norm regularization, has been widely used in imaging sciences. However, it has been found that implicit regularization outperforms explicit ones in various image processing tasks. Another issue is that the fixed explicit regularization limits the applicability to broad images since different images favor different features captured by different explicit regularizations. As such, this paper proposes a new adaptive and implicit low-rank regularization that captures the low-rank prior dynamically from the training data. The core of our new adaptive and implicit low-rank regularization is parameterizing the Laplacian matrix in the Dirichlet energy-based regularization, which we call the regularization \textit{AIR}. Theoretically, we show that the adaptive regularization of AIR enhances the implicit regularization and vanishes at the end of training. We validate AIR's effectiveness on various benchmark tasks, indicating that the AIR is particularly favorable for the scenarios when the missing entries are non-uniform. The code can be found at https://github.com/lizhemin15/AIR-Net</description><subject>Computer Science - Learning</subject><subject>Inverse problems</subject><subject>Matrix representation</subject><subject>Neural networks</subject><subject>Regularization</subject><subject>Statistics - Machine Learning</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><sourceid>GOX</sourceid><recordid>eNotj11PwjAUhhsTEwnyA7yyidfD7nT9mHeEqJAgJoT75ZS1pjjY7DpEf70DvHqTN885eR9C7lI2zrQQ7BHD0R_GkPYFU0KoKzIAztNEZwA3ZNS2W8YYSAVC8AFZTearZGnjE52U2ER_sBT3JZ3vmspvfKQr-9FVGPwvRl_v6dJ2Aas-4ncdPqmrA33DGPyRTuv-xJ6gW3LtsGrt6D-HZP3yvJ7OksX763w6WSSYC5U4w7LUZNqAhlxJiRvDlS4tyzhmRoBDRMMYz2UJnEnHnEwdSJmXpc4NKj4k95e3Z-GiCX6H4ac4iRdn8Z54uBBNqL8628ZiW3dh328qQGgAzVKu-R-j7VrM</recordid><startdate>20220627</startdate><enddate>20220627</enddate><creator>Li, Zhemin</creator><creator>Sun, Tao</creator><creator>Wang, Hongxia</creator><creator>Wang, Bao</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><scope>AKY</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20220627</creationdate><title>AIR-Net: Adaptive and Implicit Regularization Neural Network for Matrix Completion</title><author>Li, Zhemin ; Sun, Tao ; Wang, Hongxia ; Wang, Bao</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a957-fb041b48b2829766acb378de043a4b52faaab00396d2306f0f61f2669dd89ba73</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Computer Science - Learning</topic><topic>Inverse problems</topic><topic>Matrix representation</topic><topic>Neural networks</topic><topic>Regularization</topic><topic>Statistics - Machine Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Li, Zhemin</creatorcontrib><creatorcontrib>Sun, Tao</creatorcontrib><creatorcontrib>Wang, Hongxia</creatorcontrib><creatorcontrib>Wang, Bao</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><collection>arXiv Computer Science</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection><jtitle>arXiv.org</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Li, Zhemin</au><au>Sun, Tao</au><au>Wang, Hongxia</au><au>Wang, Bao</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>AIR-Net: Adaptive and Implicit Regularization Neural Network for Matrix Completion</atitle><jtitle>arXiv.org</jtitle><date>2022-06-27</date><risdate>2022</risdate><eissn>2331-8422</eissn><abstract>The explicit low-rank regularization, e.g., nuclear norm regularization, has been widely used in imaging sciences. However, it has been found that implicit regularization outperforms explicit ones in various image processing tasks. Another issue is that the fixed explicit regularization limits the applicability to broad images since different images favor different features captured by different explicit regularizations. As such, this paper proposes a new adaptive and implicit low-rank regularization that captures the low-rank prior dynamically from the training data. The core of our new adaptive and implicit low-rank regularization is parameterizing the Laplacian matrix in the Dirichlet energy-based regularization, which we call the regularization \textit{AIR}. Theoretically, we show that the adaptive regularization of AIR enhances the implicit regularization and vanishes at the end of training. We validate AIR's effectiveness on various benchmark tasks, indicating that the AIR is particularly favorable for the scenarios when the missing entries are non-uniform. The code can be found at https://github.com/lizhemin15/AIR-Net</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><doi>10.48550/arxiv.2110.07557</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2022-06
issn 2331-8422
language eng
recordid cdi_arxiv_primary_2110_07557
source arXiv.org; Free E- Journals
subjects Computer Science - Learning
Inverse problems
Matrix representation
Neural networks
Regularization
Statistics - Machine Learning
title AIR-Net: Adaptive and Implicit Regularization Neural Network for Matrix Completion
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-13T09%3A48%3A03IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_arxiv&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=AIR-Net:%20Adaptive%20and%20Implicit%20Regularization%20Neural%20Network%20for%20Matrix%20Completion&rft.jtitle=arXiv.org&rft.au=Li,%20Zhemin&rft.date=2022-06-27&rft.eissn=2331-8422&rft_id=info:doi/10.48550/arxiv.2110.07557&rft_dat=%3Cproquest_arxiv%3E2582280138%3C/proquest_arxiv%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2582280138&rft_id=info:pmid/&rfr_iscdi=true