Spectral Frank-Wolfe Algorithm: Strict Complementarity and Linear Convergence

We develop a novel variant of the classical Frank-Wolfe algorithm, which we call spectral Frank-Wolfe, for convex optimization over a spectrahedron. The spectral Frank-Wolfe algorithm has a novel ingredient: it computes a few eigenvectors of the gradient and solves a small-scale SDP in each iteratio...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ding, Lijun, Fei, Yingjie, Xu, Qiantong, Yang, Chengrun
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 Ding, Lijun
Fei, Yingjie
Xu, Qiantong
Yang, Chengrun
description We develop a novel variant of the classical Frank-Wolfe algorithm, which we call spectral Frank-Wolfe, for convex optimization over a spectrahedron. The spectral Frank-Wolfe algorithm has a novel ingredient: it computes a few eigenvectors of the gradient and solves a small-scale SDP in each iteration. Such procedure overcomes slow convergence of the classical Frank-Wolfe algorithm due to ignoring eigenvalue coalescence. We demonstrate that strict complementarity of the optimization problem is key to proving linear convergence of various algorithms, such as the spectral Frank-Wolfe algorithm as well as the projected gradient method and its accelerated version.
doi_str_mv 10.48550/arxiv.2006.01719
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2006_01719</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2006_01719</sourcerecordid><originalsourceid>FETCH-LOGICAL-a679-534ef63c83cff87699e40c3e68bc6be7b8f4fd6f525a5786fdebc0d765a9fe443</originalsourceid><addsrcrecordid>eNotz71OwzAYhWEvDKhwAUz4BhKc-r9bFVFASsXQSozRF-dzsZo4kbEqeveUwnSGVzrSQ8hDxUphpGRPkL7DqVwypkpW6creku1uRpcTDHSTIB6Lj2nwSNfDYUohf44russpuEzraZwHHDFmuIQzhdjTJkSEdEnxhOmA0eEdufEwfOH9_y7IfvO8r1-L5v3lrV43BShtC8kFesWd4c57o5W1KJjjqEznVIe6M174Xnm5lCC1Ub7HzrFeKwnWoxB8QR7_bq-edk5hhHRuf13t1cV_AJEqSZA</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Spectral Frank-Wolfe Algorithm: Strict Complementarity and Linear Convergence</title><source>arXiv.org</source><creator>Ding, Lijun ; Fei, Yingjie ; Xu, Qiantong ; Yang, Chengrun</creator><creatorcontrib>Ding, Lijun ; Fei, Yingjie ; Xu, Qiantong ; Yang, Chengrun</creatorcontrib><description>We develop a novel variant of the classical Frank-Wolfe algorithm, which we call spectral Frank-Wolfe, for convex optimization over a spectrahedron. The spectral Frank-Wolfe algorithm has a novel ingredient: it computes a few eigenvectors of the gradient and solves a small-scale SDP in each iteration. Such procedure overcomes slow convergence of the classical Frank-Wolfe algorithm due to ignoring eigenvalue coalescence. We demonstrate that strict complementarity of the optimization problem is key to proving linear convergence of various algorithms, such as the spectral Frank-Wolfe algorithm as well as the projected gradient method and its accelerated version.</description><identifier>DOI: 10.48550/arxiv.2006.01719</identifier><language>eng</language><subject>Mathematics - Optimization and Control</subject><creationdate>2020-06</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</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/2006.01719$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2006.01719$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Ding, Lijun</creatorcontrib><creatorcontrib>Fei, Yingjie</creatorcontrib><creatorcontrib>Xu, Qiantong</creatorcontrib><creatorcontrib>Yang, Chengrun</creatorcontrib><title>Spectral Frank-Wolfe Algorithm: Strict Complementarity and Linear Convergence</title><description>We develop a novel variant of the classical Frank-Wolfe algorithm, which we call spectral Frank-Wolfe, for convex optimization over a spectrahedron. The spectral Frank-Wolfe algorithm has a novel ingredient: it computes a few eigenvectors of the gradient and solves a small-scale SDP in each iteration. Such procedure overcomes slow convergence of the classical Frank-Wolfe algorithm due to ignoring eigenvalue coalescence. We demonstrate that strict complementarity of the optimization problem is key to proving linear convergence of various algorithms, such as the spectral Frank-Wolfe algorithm as well as the projected gradient method and its accelerated version.</description><subject>Mathematics - Optimization and Control</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotz71OwzAYhWEvDKhwAUz4BhKc-r9bFVFASsXQSozRF-dzsZo4kbEqeveUwnSGVzrSQ8hDxUphpGRPkL7DqVwypkpW6creku1uRpcTDHSTIB6Lj2nwSNfDYUohf44russpuEzraZwHHDFmuIQzhdjTJkSEdEnxhOmA0eEdufEwfOH9_y7IfvO8r1-L5v3lrV43BShtC8kFesWd4c57o5W1KJjjqEznVIe6M174Xnm5lCC1Ub7HzrFeKwnWoxB8QR7_bq-edk5hhHRuf13t1cV_AJEqSZA</recordid><startdate>20200602</startdate><enddate>20200602</enddate><creator>Ding, Lijun</creator><creator>Fei, Yingjie</creator><creator>Xu, Qiantong</creator><creator>Yang, Chengrun</creator><scope>AKZ</scope><scope>GOX</scope></search><sort><creationdate>20200602</creationdate><title>Spectral Frank-Wolfe Algorithm: Strict Complementarity and Linear Convergence</title><author>Ding, Lijun ; Fei, Yingjie ; Xu, Qiantong ; Yang, Chengrun</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a679-534ef63c83cff87699e40c3e68bc6be7b8f4fd6f525a5786fdebc0d765a9fe443</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Mathematics - Optimization and Control</topic><toplevel>online_resources</toplevel><creatorcontrib>Ding, Lijun</creatorcontrib><creatorcontrib>Fei, Yingjie</creatorcontrib><creatorcontrib>Xu, Qiantong</creatorcontrib><creatorcontrib>Yang, Chengrun</creatorcontrib><collection>arXiv Mathematics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Ding, Lijun</au><au>Fei, Yingjie</au><au>Xu, Qiantong</au><au>Yang, Chengrun</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Spectral Frank-Wolfe Algorithm: Strict Complementarity and Linear Convergence</atitle><date>2020-06-02</date><risdate>2020</risdate><abstract>We develop a novel variant of the classical Frank-Wolfe algorithm, which we call spectral Frank-Wolfe, for convex optimization over a spectrahedron. The spectral Frank-Wolfe algorithm has a novel ingredient: it computes a few eigenvectors of the gradient and solves a small-scale SDP in each iteration. Such procedure overcomes slow convergence of the classical Frank-Wolfe algorithm due to ignoring eigenvalue coalescence. We demonstrate that strict complementarity of the optimization problem is key to proving linear convergence of various algorithms, such as the spectral Frank-Wolfe algorithm as well as the projected gradient method and its accelerated version.</abstract><doi>10.48550/arxiv.2006.01719</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2006.01719
ispartof
issn
language eng
recordid cdi_arxiv_primary_2006_01719
source arXiv.org
subjects Mathematics - Optimization and Control
title Spectral Frank-Wolfe Algorithm: Strict Complementarity and Linear Convergence
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-31T05%3A25%3A15IST&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=Spectral%20Frank-Wolfe%20Algorithm:%20Strict%20Complementarity%20and%20Linear%20Convergence&rft.au=Ding,%20Lijun&rft.date=2020-06-02&rft_id=info:doi/10.48550/arxiv.2006.01719&rft_dat=%3Carxiv_GOX%3E2006_01719%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