Learning with risks based on M-location
In this work, we study a new class of risks defined in terms of the location and deviation of the loss distribution, generalizing far beyond classical mean-variance risk functions. The class is easily implemented as a wrapper around any smooth loss, it admits finite-sample stationarity guarantees fo...
Gespeichert in:
| Veröffentlicht in: | Machine learning 2022-12, Vol.111 (12), p.4679-4718 |
|---|---|
| 1. Verfasser: | |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 4718 |
|---|---|
| container_issue | 12 |
| container_start_page | 4679 |
| container_title | Machine learning |
| container_volume | 111 |
| creator | Holland, Matthew J. |
| description | In this work, we study a new class of risks defined in terms of the location and deviation of the loss distribution, generalizing far beyond classical mean-variance risk functions. The class is easily implemented as a wrapper around any smooth loss, it admits finite-sample stationarity guarantees for stochastic gradient methods, it is straightforward to interpret and adjust, with close links to M-estimators of the loss location, and has a salient effect on the test loss distribution, giving us control over symmetry and deviations that are not possible under naive ERM. |
| doi_str_mv | 10.1007/s10994-022-06217-5 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2741134080</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2741134080</sourcerecordid><originalsourceid>FETCH-LOGICAL-c314t-69b1a37b36baf3f5a0641034f0724d112bad59cf04d4f4e7a12a6c8900c9306e3</originalsourceid><addsrcrecordid>eNp9kEtLxDAURoMoWEf_gKuCC1fRe_NslzL4goobXYc0TcaOYzMmHcR_b7WCO1d3c8534RByinCBAPoyI9S1oMAYBcVQU7lHCpSaU5BK7pMCqkpShUwekqOc1wDAVKUKct54m4Z-WJUf_fhSpj6_5rK12XdlHMoHuonOjn0cjslBsJvsT37vgjzfXD8t72jzeHu_vGqo4yhGquoWLdctV60NPEgLSiBwEUAz0SGy1naydgFEJ4Lw2iKzylU1gKs5KM8X5Gze3ab4vvN5NOu4S8P00jAtELmACiaKzZRLMefkg9mm_s2mT4NgvoOYOYiZgpifIEZOEp-lPMHDyqe_6X-sL_XHYN0</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2741134080</pqid></control><display><type>article</type><title>Learning with risks based on M-location</title><source>SpringerLink Journals - AutoHoldings</source><creator>Holland, Matthew J.</creator><creatorcontrib>Holland, Matthew J.</creatorcontrib><description>In this work, we study a new class of risks defined in terms of the location and deviation of the loss distribution, generalizing far beyond classical mean-variance risk functions. The class is easily implemented as a wrapper around any smooth loss, it admits finite-sample stationarity guarantees for stochastic gradient methods, it is straightforward to interpret and adjust, with close links to M-estimators of the loss location, and has a salient effect on the test loss distribution, giving us control over symmetry and deviations that are not possible under naive ERM.</description><identifier>ISSN: 0885-6125</identifier><identifier>EISSN: 1573-0565</identifier><identifier>DOI: 10.1007/s10994-022-06217-5</identifier><language>eng</language><publisher>New York: Springer US</publisher><subject>Algorithms ; Artificial Intelligence ; Computer Science ; Control ; Design ; Deviation ; Machine Learning ; Mechatronics ; Natural Language Processing (NLP) ; Robotics ; Simulation and Modeling ; Special Issue of the ECML PKDD 2022 Journal Track</subject><ispartof>Machine learning, 2022-12, Vol.111 (12), p.4679-4718</ispartof><rights>The Author(s) 2022</rights><rights>The Author(s) 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><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c314t-69b1a37b36baf3f5a0641034f0724d112bad59cf04d4f4e7a12a6c8900c9306e3</cites><orcidid>0000-0002-6704-1769</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/s10994-022-06217-5$$EPDF$$P50$$Gspringer$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://link.springer.com/10.1007/s10994-022-06217-5$$EHTML$$P50$$Gspringer$$Hfree_for_read</linktohtml><link.rule.ids>314,777,781,27905,27906,41469,42538,51300</link.rule.ids></links><search><creatorcontrib>Holland, Matthew J.</creatorcontrib><title>Learning with risks based on M-location</title><title>Machine learning</title><addtitle>Mach Learn</addtitle><description>In this work, we study a new class of risks defined in terms of the location and deviation of the loss distribution, generalizing far beyond classical mean-variance risk functions. The class is easily implemented as a wrapper around any smooth loss, it admits finite-sample stationarity guarantees for stochastic gradient methods, it is straightforward to interpret and adjust, with close links to M-estimators of the loss location, and has a salient effect on the test loss distribution, giving us control over symmetry and deviations that are not possible under naive ERM.</description><subject>Algorithms</subject><subject>Artificial Intelligence</subject><subject>Computer Science</subject><subject>Control</subject><subject>Design</subject><subject>Deviation</subject><subject>Machine Learning</subject><subject>Mechatronics</subject><subject>Natural Language Processing (NLP)</subject><subject>Robotics</subject><subject>Simulation and Modeling</subject><subject>Special Issue of the ECML PKDD 2022 Journal Track</subject><issn>0885-6125</issn><issn>1573-0565</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>C6C</sourceid><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><sourceid>GNUQQ</sourceid><recordid>eNp9kEtLxDAURoMoWEf_gKuCC1fRe_NslzL4goobXYc0TcaOYzMmHcR_b7WCO1d3c8534RByinCBAPoyI9S1oMAYBcVQU7lHCpSaU5BK7pMCqkpShUwekqOc1wDAVKUKct54m4Z-WJUf_fhSpj6_5rK12XdlHMoHuonOjn0cjslBsJvsT37vgjzfXD8t72jzeHu_vGqo4yhGquoWLdctV60NPEgLSiBwEUAz0SGy1naydgFEJ4Lw2iKzylU1gKs5KM8X5Gze3ab4vvN5NOu4S8P00jAtELmACiaKzZRLMefkg9mm_s2mT4NgvoOYOYiZgpifIEZOEp-lPMHDyqe_6X-sL_XHYN0</recordid><startdate>20221201</startdate><enddate>20221201</enddate><creator>Holland, Matthew J.</creator><general>Springer US</general><general>Springer Nature B.V</general><scope>C6C</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>3V.</scope><scope>7SC</scope><scope>7XB</scope><scope>88I</scope><scope>8AL</scope><scope>8AO</scope><scope>8FD</scope><scope>8FE</scope><scope>8FG</scope><scope>8FK</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>ARAPS</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>GNUQQ</scope><scope>HCIFZ</scope><scope>JQ2</scope><scope>K7-</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>M0N</scope><scope>M2P</scope><scope>P5Z</scope><scope>P62</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>Q9U</scope><orcidid>https://orcid.org/0000-0002-6704-1769</orcidid></search><sort><creationdate>20221201</creationdate><title>Learning with risks based on M-location</title><author>Holland, Matthew J.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c314t-69b1a37b36baf3f5a0641034f0724d112bad59cf04d4f4e7a12a6c8900c9306e3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Algorithms</topic><topic>Artificial Intelligence</topic><topic>Computer Science</topic><topic>Control</topic><topic>Design</topic><topic>Deviation</topic><topic>Machine Learning</topic><topic>Mechatronics</topic><topic>Natural Language Processing (NLP)</topic><topic>Robotics</topic><topic>Simulation and Modeling</topic><topic>Special Issue of the ECML PKDD 2022 Journal Track</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Holland, Matthew J.</creatorcontrib><collection>Springer Nature OA Free Journals</collection><collection>CrossRef</collection><collection>ProQuest Central (Corporate)</collection><collection>Computer and Information Systems Abstracts</collection><collection>ProQuest Central (purchase pre-March 2016)</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>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>Advanced Technologies & Aerospace Collection</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>ProQuest Central Student</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Computer Science Collection</collection><collection>Computer Science Database</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>Computing Database</collection><collection>Science Database</collection><collection>Advanced Technologies & Aerospace Database</collection><collection>ProQuest Advanced Technologies & Aerospace Collection</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>ProQuest Central Basic</collection><jtitle>Machine learning</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Holland, Matthew J.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Learning with risks based on M-location</atitle><jtitle>Machine learning</jtitle><stitle>Mach Learn</stitle><date>2022-12-01</date><risdate>2022</risdate><volume>111</volume><issue>12</issue><spage>4679</spage><epage>4718</epage><pages>4679-4718</pages><issn>0885-6125</issn><eissn>1573-0565</eissn><abstract>In this work, we study a new class of risks defined in terms of the location and deviation of the loss distribution, generalizing far beyond classical mean-variance risk functions. The class is easily implemented as a wrapper around any smooth loss, it admits finite-sample stationarity guarantees for stochastic gradient methods, it is straightforward to interpret and adjust, with close links to M-estimators of the loss location, and has a salient effect on the test loss distribution, giving us control over symmetry and deviations that are not possible under naive ERM.</abstract><cop>New York</cop><pub>Springer US</pub><doi>10.1007/s10994-022-06217-5</doi><tpages>40</tpages><orcidid>https://orcid.org/0000-0002-6704-1769</orcidid><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0885-6125 |
| ispartof | Machine learning, 2022-12, Vol.111 (12), p.4679-4718 |
| issn | 0885-6125 1573-0565 |
| language | eng |
| recordid | cdi_proquest_journals_2741134080 |
| source | SpringerLink Journals - AutoHoldings |
| subjects | Algorithms Artificial Intelligence Computer Science Control Design Deviation Machine Learning Mechatronics Natural Language Processing (NLP) Robotics Simulation and Modeling Special Issue of the ECML PKDD 2022 Journal Track |
| title | Learning with risks based on M-location |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-18T21%3A07%3A15IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Learning%20with%20risks%20based%20on%20M-location&rft.jtitle=Machine%20learning&rft.au=Holland,%20Matthew%20J.&rft.date=2022-12-01&rft.volume=111&rft.issue=12&rft.spage=4679&rft.epage=4718&rft.pages=4679-4718&rft.issn=0885-6125&rft.eissn=1573-0565&rft_id=info:doi/10.1007/s10994-022-06217-5&rft_dat=%3Cproquest_cross%3E2741134080%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2741134080&rft_id=info:pmid/&rfr_iscdi=true |