An inexact successive quadratic approximation method for L-1 regularized optimization

We study a Newton-like method for the minimization of an objective function ϕ that is the sum of a smooth function and an ℓ 1 regularization term. This method, which is sometimes referred to in the literature as a proximal Newton method, computes a step by minimizing a piecewise quadratic model q k...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Mathematical programming 2016-06, Vol.157 (2), p.375-396
Hauptverfasser: Byrd, Richard H., Nocedal, Jorge, Oztoprak, Figen
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 396
container_issue 2
container_start_page 375
container_title Mathematical programming
container_volume 157
creator Byrd, Richard H.
Nocedal, Jorge
Oztoprak, Figen
description We study a Newton-like method for the minimization of an objective function ϕ that is the sum of a smooth function and an ℓ 1 regularization term. This method, which is sometimes referred to in the literature as a proximal Newton method, computes a step by minimizing a piecewise quadratic model q k of the objective function ϕ . In order to make this approach efficient in practice, it is imperative to perform this inner minimization inexactly. In this paper, we give inexactness conditions that guarantee global convergence and that can be used to control the local rate of convergence of the iteration. Our inexactness conditions are based on a semi-smooth function that represents a (continuous) measure of the optimality conditions of the problem, and that embodies the soft-thresholding iteration. We give careful consideration to the algorithm employed for the inner minimization, and report numerical results on two test sets originating in machine learning.
doi_str_mv 10.1007/s10107-015-0941-y
format Article
fullrecord <record><control><sourceid>proquest_sprin</sourceid><recordid>TN_cdi_proquest_miscellaneous_1825468261</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1825468261</sourcerecordid><originalsourceid>FETCH-LOGICAL-p189t-df01da09810a8fb253497f3ab2d82313a5aca767b7873c635488ced3026dc97d3</originalsourceid><addsrcrecordid>eNpd0E1LAzEQgOEgCtbqD_AW8OIlOpPsJtljKX5BwYs9L2mSrSn71c2utP31bq0H8TQMPAzDS8gtwgMCqMeIgKAYYMogS5Dtz8gEEyFZIhN5TiYAPGWpRLgkVzFuAACF1hOynNU01H5nbE_jYK2PMXx5uh2M60wfLDVt2zW7UI1LU9PK95-No0XT0QVD2vn1UJouHLyjTduHKhx-3DW5KEwZ_c3vnJLl89PH_JUt3l_e5rMFa1FnPXMFoDOQaQSjixVPRZKpQpgVd5oLFCY11iipVkorYaVIE62tdwK4dDZTTkzJ_enu-ON28LHPqxCtL0tT-2aIOWqeJlJziSO9-0c3zdDV43c5qgy5gESmo-InFdsu1Gvf_VGQH0vnp9L5WDo_ls734hvNo3Ga</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1791230465</pqid></control><display><type>article</type><title>An inexact successive quadratic approximation method for L-1 regularized optimization</title><source>SpringerNature Journals</source><source>EBSCOhost Business Source Complete</source><creator>Byrd, Richard H. ; Nocedal, Jorge ; Oztoprak, Figen</creator><creatorcontrib>Byrd, Richard H. ; Nocedal, Jorge ; Oztoprak, Figen</creatorcontrib><description>We study a Newton-like method for the minimization of an objective function ϕ that is the sum of a smooth function and an ℓ 1 regularization term. This method, which is sometimes referred to in the literature as a proximal Newton method, computes a step by minimizing a piecewise quadratic model q k of the objective function ϕ . In order to make this approach efficient in practice, it is imperative to perform this inner minimization inexactly. In this paper, we give inexactness conditions that guarantee global convergence and that can be used to control the local rate of convergence of the iteration. Our inexactness conditions are based on a semi-smooth function that represents a (continuous) measure of the optimality conditions of the problem, and that embodies the soft-thresholding iteration. We give careful consideration to the algorithm employed for the inner minimization, and report numerical results on two test sets originating in machine learning.</description><identifier>ISSN: 0025-5610</identifier><identifier>EISSN: 1436-4646</identifier><identifier>DOI: 10.1007/s10107-015-0941-y</identifier><identifier>CODEN: MHPGA4</identifier><language>eng</language><publisher>Berlin/Heidelberg: Springer Berlin Heidelberg</publisher><subject>Algorithms ; Approximation ; Artificial intelligence ; Calculus of Variations and Optimal Control; Optimization ; Combinatorics ; Convergence ; Full Length Paper ; Industrial engineering ; Iterative methods ; Mathematical analysis ; Mathematical and Computational Physics ; Mathematical Methods in Physics ; Mathematical models ; Mathematics ; Mathematics and Statistics ; Mathematics of Computing ; Minimization ; Newton methods ; Numerical Analysis ; Optimization ; Studies ; Test sets ; Theoretical</subject><ispartof>Mathematical programming, 2016-06, Vol.157 (2), p.375-396</ispartof><rights>Springer-Verlag Berlin Heidelberg and Mathematical Optimization Society 2015</rights><rights>Springer-Verlag Berlin Heidelberg and Mathematical Optimization Society 2016</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/s10107-015-0941-y$$EPDF$$P50$$Gspringer$$H</linktopdf><linktohtml>$$Uhttps://link.springer.com/10.1007/s10107-015-0941-y$$EHTML$$P50$$Gspringer$$H</linktohtml><link.rule.ids>314,780,784,27924,27925,41488,42557,51319</link.rule.ids></links><search><creatorcontrib>Byrd, Richard H.</creatorcontrib><creatorcontrib>Nocedal, Jorge</creatorcontrib><creatorcontrib>Oztoprak, Figen</creatorcontrib><title>An inexact successive quadratic approximation method for L-1 regularized optimization</title><title>Mathematical programming</title><addtitle>Math. Program</addtitle><description>We study a Newton-like method for the minimization of an objective function ϕ that is the sum of a smooth function and an ℓ 1 regularization term. This method, which is sometimes referred to in the literature as a proximal Newton method, computes a step by minimizing a piecewise quadratic model q k of the objective function ϕ . In order to make this approach efficient in practice, it is imperative to perform this inner minimization inexactly. In this paper, we give inexactness conditions that guarantee global convergence and that can be used to control the local rate of convergence of the iteration. Our inexactness conditions are based on a semi-smooth function that represents a (continuous) measure of the optimality conditions of the problem, and that embodies the soft-thresholding iteration. We give careful consideration to the algorithm employed for the inner minimization, and report numerical results on two test sets originating in machine learning.</description><subject>Algorithms</subject><subject>Approximation</subject><subject>Artificial intelligence</subject><subject>Calculus of Variations and Optimal Control; Optimization</subject><subject>Combinatorics</subject><subject>Convergence</subject><subject>Full Length Paper</subject><subject>Industrial engineering</subject><subject>Iterative methods</subject><subject>Mathematical analysis</subject><subject>Mathematical and Computational Physics</subject><subject>Mathematical Methods in Physics</subject><subject>Mathematical models</subject><subject>Mathematics</subject><subject>Mathematics and Statistics</subject><subject>Mathematics of Computing</subject><subject>Minimization</subject><subject>Newton methods</subject><subject>Numerical Analysis</subject><subject>Optimization</subject><subject>Studies</subject><subject>Test sets</subject><subject>Theoretical</subject><issn>0025-5610</issn><issn>1436-4646</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2016</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><sourceid>GNUQQ</sourceid><recordid>eNpd0E1LAzEQgOEgCtbqD_AW8OIlOpPsJtljKX5BwYs9L2mSrSn71c2utP31bq0H8TQMPAzDS8gtwgMCqMeIgKAYYMogS5Dtz8gEEyFZIhN5TiYAPGWpRLgkVzFuAACF1hOynNU01H5nbE_jYK2PMXx5uh2M60wfLDVt2zW7UI1LU9PK95-No0XT0QVD2vn1UJouHLyjTduHKhx-3DW5KEwZ_c3vnJLl89PH_JUt3l_e5rMFa1FnPXMFoDOQaQSjixVPRZKpQpgVd5oLFCY11iipVkorYaVIE62tdwK4dDZTTkzJ_enu-ON28LHPqxCtL0tT-2aIOWqeJlJziSO9-0c3zdDV43c5qgy5gESmo-InFdsu1Gvf_VGQH0vnp9L5WDo_ls734hvNo3Ga</recordid><startdate>20160601</startdate><enddate>20160601</enddate><creator>Byrd, Richard H.</creator><creator>Nocedal, Jorge</creator><creator>Oztoprak, Figen</creator><general>Springer Berlin Heidelberg</general><general>Springer Nature B.V</general><scope>3V.</scope><scope>7SC</scope><scope>7WY</scope><scope>7WZ</scope><scope>7XB</scope><scope>87Z</scope><scope>88I</scope><scope>8AL</scope><scope>8AO</scope><scope>8FD</scope><scope>8FE</scope><scope>8FG</scope><scope>8FK</scope><scope>8FL</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>ARAPS</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BEZIV</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>FRNLG</scope><scope>F~G</scope><scope>GNUQQ</scope><scope>HCIFZ</scope><scope>JQ2</scope><scope>K60</scope><scope>K6~</scope><scope>K7-</scope><scope>L.-</scope><scope>L6V</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>M0C</scope><scope>M0N</scope><scope>M2P</scope><scope>M7S</scope><scope>P5Z</scope><scope>P62</scope><scope>PQBIZ</scope><scope>PQBZA</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PTHSS</scope><scope>Q9U</scope></search><sort><creationdate>20160601</creationdate><title>An inexact successive quadratic approximation method for L-1 regularized optimization</title><author>Byrd, Richard H. ; Nocedal, Jorge ; Oztoprak, Figen</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-p189t-df01da09810a8fb253497f3ab2d82313a5aca767b7873c635488ced3026dc97d3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2016</creationdate><topic>Algorithms</topic><topic>Approximation</topic><topic>Artificial intelligence</topic><topic>Calculus of Variations and Optimal Control; Optimization</topic><topic>Combinatorics</topic><topic>Convergence</topic><topic>Full Length Paper</topic><topic>Industrial engineering</topic><topic>Iterative methods</topic><topic>Mathematical analysis</topic><topic>Mathematical and Computational Physics</topic><topic>Mathematical Methods in Physics</topic><topic>Mathematical models</topic><topic>Mathematics</topic><topic>Mathematics and Statistics</topic><topic>Mathematics of Computing</topic><topic>Minimization</topic><topic>Newton methods</topic><topic>Numerical Analysis</topic><topic>Optimization</topic><topic>Studies</topic><topic>Test sets</topic><topic>Theoretical</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Byrd, Richard H.</creatorcontrib><creatorcontrib>Nocedal, Jorge</creatorcontrib><creatorcontrib>Oztoprak, Figen</creatorcontrib><collection>ProQuest Central (Corporate)</collection><collection>Computer and Information Systems Abstracts</collection><collection>Access via ABI/INFORM (ProQuest)</collection><collection>ABI/INFORM Global (PDF only)</collection><collection>ProQuest Central (purchase pre-March 2016)</collection><collection>ABI/INFORM Global (Alumni Edition)</collection><collection>Science Database (Alumni Edition)</collection><collection>Computing Database (Alumni Edition)</collection><collection>ProQuest Pharma Collection</collection><collection>Technology Research Database</collection><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>ProQuest Central (Alumni) (purchase pre-March 2016)</collection><collection>ABI/INFORM Collection (Alumni Edition)</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>Advanced Technologies &amp; Aerospace Collection</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Business Premium Collection</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>Business Premium Collection (Alumni)</collection><collection>ABI/INFORM Global (Corporate)</collection><collection>ProQuest Central Student</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Computer Science Collection</collection><collection>ProQuest Business Collection (Alumni Edition)</collection><collection>ProQuest Business Collection</collection><collection>Computer Science Database</collection><collection>ABI/INFORM Professional Advanced</collection><collection>ProQuest Engineering Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts – Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><collection>ABI/INFORM Global</collection><collection>Computing Database</collection><collection>Science Database</collection><collection>Engineering Database</collection><collection>Advanced Technologies &amp; Aerospace Database</collection><collection>ProQuest Advanced Technologies &amp; Aerospace Collection</collection><collection>ProQuest One Business</collection><collection>ProQuest One Business (Alumni)</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>Engineering Collection</collection><collection>ProQuest Central Basic</collection><jtitle>Mathematical programming</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Byrd, Richard H.</au><au>Nocedal, Jorge</au><au>Oztoprak, Figen</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>An inexact successive quadratic approximation method for L-1 regularized optimization</atitle><jtitle>Mathematical programming</jtitle><stitle>Math. Program</stitle><date>2016-06-01</date><risdate>2016</risdate><volume>157</volume><issue>2</issue><spage>375</spage><epage>396</epage><pages>375-396</pages><issn>0025-5610</issn><eissn>1436-4646</eissn><coden>MHPGA4</coden><abstract>We study a Newton-like method for the minimization of an objective function ϕ that is the sum of a smooth function and an ℓ 1 regularization term. This method, which is sometimes referred to in the literature as a proximal Newton method, computes a step by minimizing a piecewise quadratic model q k of the objective function ϕ . In order to make this approach efficient in practice, it is imperative to perform this inner minimization inexactly. In this paper, we give inexactness conditions that guarantee global convergence and that can be used to control the local rate of convergence of the iteration. Our inexactness conditions are based on a semi-smooth function that represents a (continuous) measure of the optimality conditions of the problem, and that embodies the soft-thresholding iteration. We give careful consideration to the algorithm employed for the inner minimization, and report numerical results on two test sets originating in machine learning.</abstract><cop>Berlin/Heidelberg</cop><pub>Springer Berlin Heidelberg</pub><doi>10.1007/s10107-015-0941-y</doi><tpages>22</tpages></addata></record>
fulltext fulltext
identifier ISSN: 0025-5610
ispartof Mathematical programming, 2016-06, Vol.157 (2), p.375-396
issn 0025-5610
1436-4646
language eng
recordid cdi_proquest_miscellaneous_1825468261
source SpringerNature Journals; EBSCOhost Business Source Complete
subjects Algorithms
Approximation
Artificial intelligence
Calculus of Variations and Optimal Control
Optimization
Combinatorics
Convergence
Full Length Paper
Industrial engineering
Iterative methods
Mathematical analysis
Mathematical and Computational Physics
Mathematical Methods in Physics
Mathematical models
Mathematics
Mathematics and Statistics
Mathematics of Computing
Minimization
Newton methods
Numerical Analysis
Optimization
Studies
Test sets
Theoretical
title An inexact successive quadratic approximation method for L-1 regularized optimization
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T02%3A58%3A08IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_sprin&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=An%20inexact%20successive%20quadratic%20approximation%20method%20for%20L-1%20regularized%20optimization&rft.jtitle=Mathematical%20programming&rft.au=Byrd,%20Richard%20H.&rft.date=2016-06-01&rft.volume=157&rft.issue=2&rft.spage=375&rft.epage=396&rft.pages=375-396&rft.issn=0025-5610&rft.eissn=1436-4646&rft.coden=MHPGA4&rft_id=info:doi/10.1007/s10107-015-0941-y&rft_dat=%3Cproquest_sprin%3E1825468261%3C/proquest_sprin%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=1791230465&rft_id=info:pmid/&rfr_iscdi=true