A Class of High Order Tuners for Adaptive Systems
Parameter estimation algorithms using higher order gradient-based methods are increasingly sought after in machine learning. Such methods however, may become unstable when regressors are time-varying. Inspired by techniques employed in adaptive systems, this letter proposes a new variational perspec...
Gespeichert in:
| Veröffentlicht in: | IEEE control systems letters 2021-04, Vol.5 (2), p.391-396 |
|---|---|
| Hauptverfasser: | , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 396 |
|---|---|
| container_issue | 2 |
| container_start_page | 391 |
| container_title | IEEE control systems letters |
| container_volume | 5 |
| creator | Gaudio, Joseph E. Annaswamy, Anuradha M. Bolender, Michael A. Lavretsky, Eugene Gibson, Travis E. |
| description | Parameter estimation algorithms using higher order gradient-based methods are increasingly sought after in machine learning. Such methods however, may become unstable when regressors are time-varying. Inspired by techniques employed in adaptive systems, this letter proposes a new variational perspective to derive four higher order tuners with provable stability guarantees. This perspective includes concepts based on higher order tuners and normalization and allows stability to be established for problems with time-varying regressors. The stability analysis builds on a novel technique which stems from symplectic mechanics, that links Lagrangians and Hamiltonians to the underlying Lyapunov stability analysis, and is provided for common linear-in-parameter models. |
| doi_str_mv | 10.1109/LCSYS.2020.3002513 |
| format | Article |
| fullrecord | <record><control><sourceid>crossref_RIE</sourceid><recordid>TN_cdi_crossref_primary_10_1109_LCSYS_2020_3002513</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>9119094</ieee_id><sourcerecordid>10_1109_LCSYS_2020_3002513</sourcerecordid><originalsourceid>FETCH-LOGICAL-c267t-ec7ceecf93152e748a376cf934ca8740e5209a637824a87611965a60eeb2125e3</originalsourceid><addsrcrecordid>eNpNj81Kw0AUhQdRsNS-gG7mBRLv3PnLLENQWwh0kbpwFcb0RiOtKTNR6Nub2CKuzj0XvgMfY7cCUiHA3ZdF9VKlCAipBEAt5AWbobI6EUqby3_3NVvE-AEAIkML6GZM5LzY-Rh53_Jl9_bO12FLgW--PilE3vaB51t_GLpv4tUxDrSPN-yq9btIi3PO2fPjw6ZYJuX6aVXkZdKgsUNCjW2ImtZJoZGsyry0Zqqq8ZlVQBrBeSNthmp8GCGc0d4A0SsK1CTnDE-7TehjDNTWh9DtfTjWAurJu_71rifv-uw9QncnqCOiP8CN4-CU_AGzXFHt</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>A Class of High Order Tuners for Adaptive Systems</title><source>IEEE Electronic Library (IEL)</source><creator>Gaudio, Joseph E. ; Annaswamy, Anuradha M. ; Bolender, Michael A. ; Lavretsky, Eugene ; Gibson, Travis E.</creator><creatorcontrib>Gaudio, Joseph E. ; Annaswamy, Anuradha M. ; Bolender, Michael A. ; Lavretsky, Eugene ; Gibson, Travis E.</creatorcontrib><description>Parameter estimation algorithms using higher order gradient-based methods are increasingly sought after in machine learning. Such methods however, may become unstable when regressors are time-varying. Inspired by techniques employed in adaptive systems, this letter proposes a new variational perspective to derive four higher order tuners with provable stability guarantees. This perspective includes concepts based on higher order tuners and normalization and allows stability to be established for problems with time-varying regressors. The stability analysis builds on a novel technique which stems from symplectic mechanics, that links Lagrangians and Hamiltonians to the underlying Lyapunov stability analysis, and is provided for common linear-in-parameter models.</description><identifier>ISSN: 2475-1456</identifier><identifier>EISSN: 2475-1456</identifier><identifier>DOI: 10.1109/LCSYS.2020.3002513</identifier><identifier>CODEN: ICSLBO</identifier><language>eng</language><publisher>IEEE</publisher><subject>Adaptive systems ; Differential equations ; Lyapunov methods ; Machine learning ; Machine learning algorithms ; Stability analysis ; Tuners ; uncertain systems</subject><ispartof>IEEE control systems letters, 2021-04, Vol.5 (2), p.391-396</ispartof><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c267t-ec7ceecf93152e748a376cf934ca8740e5209a637824a87611965a60eeb2125e3</citedby><cites>FETCH-LOGICAL-c267t-ec7ceecf93152e748a376cf934ca8740e5209a637824a87611965a60eeb2125e3</cites><orcidid>0000-0003-0443-786X ; 0000-0003-0320-8010 ; 0000-0002-4354-0459 ; 0000-0002-9557-0414 ; 0000-0001-5471-6571</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/9119094$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>314,778,782,794,27911,27912,54745</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/9119094$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Gaudio, Joseph E.</creatorcontrib><creatorcontrib>Annaswamy, Anuradha M.</creatorcontrib><creatorcontrib>Bolender, Michael A.</creatorcontrib><creatorcontrib>Lavretsky, Eugene</creatorcontrib><creatorcontrib>Gibson, Travis E.</creatorcontrib><title>A Class of High Order Tuners for Adaptive Systems</title><title>IEEE control systems letters</title><addtitle>LCSYS</addtitle><description>Parameter estimation algorithms using higher order gradient-based methods are increasingly sought after in machine learning. Such methods however, may become unstable when regressors are time-varying. Inspired by techniques employed in adaptive systems, this letter proposes a new variational perspective to derive four higher order tuners with provable stability guarantees. This perspective includes concepts based on higher order tuners and normalization and allows stability to be established for problems with time-varying regressors. The stability analysis builds on a novel technique which stems from symplectic mechanics, that links Lagrangians and Hamiltonians to the underlying Lyapunov stability analysis, and is provided for common linear-in-parameter models.</description><subject>Adaptive systems</subject><subject>Differential equations</subject><subject>Lyapunov methods</subject><subject>Machine learning</subject><subject>Machine learning algorithms</subject><subject>Stability analysis</subject><subject>Tuners</subject><subject>uncertain systems</subject><issn>2475-1456</issn><issn>2475-1456</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>RIE</sourceid><recordid>eNpNj81Kw0AUhQdRsNS-gG7mBRLv3PnLLENQWwh0kbpwFcb0RiOtKTNR6Nub2CKuzj0XvgMfY7cCUiHA3ZdF9VKlCAipBEAt5AWbobI6EUqby3_3NVvE-AEAIkML6GZM5LzY-Rh53_Jl9_bO12FLgW--PilE3vaB51t_GLpv4tUxDrSPN-yq9btIi3PO2fPjw6ZYJuX6aVXkZdKgsUNCjW2ImtZJoZGsyry0Zqqq8ZlVQBrBeSNthmp8GCGc0d4A0SsK1CTnDE-7TehjDNTWh9DtfTjWAurJu_71rifv-uw9QncnqCOiP8CN4-CU_AGzXFHt</recordid><startdate>202104</startdate><enddate>202104</enddate><creator>Gaudio, Joseph E.</creator><creator>Annaswamy, Anuradha M.</creator><creator>Bolender, Michael A.</creator><creator>Lavretsky, Eugene</creator><creator>Gibson, Travis E.</creator><general>IEEE</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><orcidid>https://orcid.org/0000-0003-0443-786X</orcidid><orcidid>https://orcid.org/0000-0003-0320-8010</orcidid><orcidid>https://orcid.org/0000-0002-4354-0459</orcidid><orcidid>https://orcid.org/0000-0002-9557-0414</orcidid><orcidid>https://orcid.org/0000-0001-5471-6571</orcidid></search><sort><creationdate>202104</creationdate><title>A Class of High Order Tuners for Adaptive Systems</title><author>Gaudio, Joseph E. ; Annaswamy, Anuradha M. ; Bolender, Michael A. ; Lavretsky, Eugene ; Gibson, Travis E.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c267t-ec7ceecf93152e748a376cf934ca8740e5209a637824a87611965a60eeb2125e3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Adaptive systems</topic><topic>Differential equations</topic><topic>Lyapunov methods</topic><topic>Machine learning</topic><topic>Machine learning algorithms</topic><topic>Stability analysis</topic><topic>Tuners</topic><topic>uncertain systems</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Gaudio, Joseph E.</creatorcontrib><creatorcontrib>Annaswamy, Anuradha M.</creatorcontrib><creatorcontrib>Bolender, Michael A.</creatorcontrib><creatorcontrib>Lavretsky, Eugene</creatorcontrib><creatorcontrib>Gibson, Travis E.</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998-Present</collection><collection>IEEE Electronic Library (IEL)</collection><collection>CrossRef</collection><jtitle>IEEE control systems letters</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Gaudio, Joseph E.</au><au>Annaswamy, Anuradha M.</au><au>Bolender, Michael A.</au><au>Lavretsky, Eugene</au><au>Gibson, Travis E.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A Class of High Order Tuners for Adaptive Systems</atitle><jtitle>IEEE control systems letters</jtitle><stitle>LCSYS</stitle><date>2021-04</date><risdate>2021</risdate><volume>5</volume><issue>2</issue><spage>391</spage><epage>396</epage><pages>391-396</pages><issn>2475-1456</issn><eissn>2475-1456</eissn><coden>ICSLBO</coden><abstract>Parameter estimation algorithms using higher order gradient-based methods are increasingly sought after in machine learning. Such methods however, may become unstable when regressors are time-varying. Inspired by techniques employed in adaptive systems, this letter proposes a new variational perspective to derive four higher order tuners with provable stability guarantees. This perspective includes concepts based on higher order tuners and normalization and allows stability to be established for problems with time-varying regressors. The stability analysis builds on a novel technique which stems from symplectic mechanics, that links Lagrangians and Hamiltonians to the underlying Lyapunov stability analysis, and is provided for common linear-in-parameter models.</abstract><pub>IEEE</pub><doi>10.1109/LCSYS.2020.3002513</doi><tpages>6</tpages><orcidid>https://orcid.org/0000-0003-0443-786X</orcidid><orcidid>https://orcid.org/0000-0003-0320-8010</orcidid><orcidid>https://orcid.org/0000-0002-4354-0459</orcidid><orcidid>https://orcid.org/0000-0002-9557-0414</orcidid><orcidid>https://orcid.org/0000-0001-5471-6571</orcidid></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 2475-1456 |
| ispartof | IEEE control systems letters, 2021-04, Vol.5 (2), p.391-396 |
| issn | 2475-1456 2475-1456 |
| language | eng |
| recordid | cdi_crossref_primary_10_1109_LCSYS_2020_3002513 |
| source | IEEE Electronic Library (IEL) |
| subjects | Adaptive systems Differential equations Lyapunov methods Machine learning Machine learning algorithms Stability analysis Tuners uncertain systems |
| title | A Class of High Order Tuners for Adaptive Systems |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-15T15%3A03%3A58IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-crossref_RIE&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20Class%20of%20High%20Order%20Tuners%20for%20Adaptive%20Systems&rft.jtitle=IEEE%20control%20systems%20letters&rft.au=Gaudio,%20Joseph%20E.&rft.date=2021-04&rft.volume=5&rft.issue=2&rft.spage=391&rft.epage=396&rft.pages=391-396&rft.issn=2475-1456&rft.eissn=2475-1456&rft.coden=ICSLBO&rft_id=info:doi/10.1109/LCSYS.2020.3002513&rft_dat=%3Ccrossref_RIE%3E10_1109_LCSYS_2020_3002513%3C/crossref_RIE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=9119094&rfr_iscdi=true |