Semi-supervised geometric mean of Kullback-Leibler divergences for subspace selection
Subspace selection is widely adopted in many areas of pattern recognition. A recent result, named maximizing the geometric mean of Kullback-Leibler (KL) divergences of class pairs (MGMD), is a successful method for subspace selection, which can significantly reduce the class separation problem. Howe...
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Tagungsbericht |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 1235 |
|---|---|
| container_issue | |
| container_start_page | 1232 |
| container_title | |
| container_volume | 2 |
| creator | Si-Bao Chen Hai-Xian Wang Xing-Yi Zhang Bin Luo |
| description | Subspace selection is widely adopted in many areas of pattern recognition. A recent result, named maximizing the geometric mean of Kullback-Leibler (KL) divergences of class pairs (MGMD), is a successful method for subspace selection, which can significantly reduce the class separation problem. However, in many applications, labeled data are very limited while unlabeled data can be easily obtained. The estimation of divergences of class pairs is unstable using inadequate labeled data. To take advantage of unlabeled data for subspace selection, semi-supervised MGMD (SSMGMD) is proposed using graph Laplacian as normalization. Quasi-Newton method is adopted to solve the optimization problem. Experiments on synthetic data and real image data show the validity of SSMGMD. |
| doi_str_mv | 10.1109/FSKD.2011.6019712 |
| format | Conference Proceeding |
| fullrecord | <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_6019712</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>6019712</ieee_id><sourcerecordid>6019712</sourcerecordid><originalsourceid>FETCH-LOGICAL-i90t-40893bf47f9f0f0260b8ba3846febf546397072959951ab95157aba7d8da263e3</originalsourceid><addsrcrecordid>eNpVkL1OwzAUhY0QEqjkARCLXyDBjhPbd0SFUtRIDC1zZSfXlSF_spNKvD2R6MIZzqdvOcMh5IGzjHMGT5v97iXLGeeZZBwUz69IAkpzyXNd8IXX_5zBLUli_GJLpASh9B353GPn0ziPGM4-YkNPOHQ4BV_TDk1PB0d3c9taU3-nFXrbYqCNP2M4YV9jpG4INM42jqZGGrHFevJDf09unGkjJheuyGHzelhv0-rj7X39XKUe2JQWTIOwrlAOHHMsl8xqa4QupEPrykIKUEzlUAKU3NilSmWsUY1uTC4FihV5_Jv1iHgcg-9M-DlevhC_5_1Smw</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Semi-supervised geometric mean of Kullback-Leibler divergences for subspace selection</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Si-Bao Chen ; Hai-Xian Wang ; Xing-Yi Zhang ; Bin Luo</creator><creatorcontrib>Si-Bao Chen ; Hai-Xian Wang ; Xing-Yi Zhang ; Bin Luo</creatorcontrib><description>Subspace selection is widely adopted in many areas of pattern recognition. A recent result, named maximizing the geometric mean of Kullback-Leibler (KL) divergences of class pairs (MGMD), is a successful method for subspace selection, which can significantly reduce the class separation problem. However, in many applications, labeled data are very limited while unlabeled data can be easily obtained. The estimation of divergences of class pairs is unstable using inadequate labeled data. To take advantage of unlabeled data for subspace selection, semi-supervised MGMD (SSMGMD) is proposed using graph Laplacian as normalization. Quasi-Newton method is adopted to solve the optimization problem. Experiments on synthetic data and real image data show the validity of SSMGMD.</description><identifier>ISBN: 9781612841809</identifier><identifier>ISBN: 1612841805</identifier><identifier>EISBN: 9781612841816</identifier><identifier>EISBN: 1612841813</identifier><identifier>EISBN: 1612841791</identifier><identifier>EISBN: 9781612841793</identifier><identifier>DOI: 10.1109/FSKD.2011.6019712</identifier><language>eng</language><publisher>IEEE</publisher><subject>Covariance matrix ; Educational institutions ; Laplace equations ; Manifolds ; Optimization ; Symmetric matrices ; Training</subject><ispartof>2011 Eighth International Conference on Fuzzy Systems and Knowledge Discovery (FSKD), 2011, Vol.2, p.1232-1235</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/6019712$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,776,780,785,786,2052,27904,54898</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/6019712$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Si-Bao Chen</creatorcontrib><creatorcontrib>Hai-Xian Wang</creatorcontrib><creatorcontrib>Xing-Yi Zhang</creatorcontrib><creatorcontrib>Bin Luo</creatorcontrib><title>Semi-supervised geometric mean of Kullback-Leibler divergences for subspace selection</title><title>2011 Eighth International Conference on Fuzzy Systems and Knowledge Discovery (FSKD)</title><addtitle>FSKD</addtitle><description>Subspace selection is widely adopted in many areas of pattern recognition. A recent result, named maximizing the geometric mean of Kullback-Leibler (KL) divergences of class pairs (MGMD), is a successful method for subspace selection, which can significantly reduce the class separation problem. However, in many applications, labeled data are very limited while unlabeled data can be easily obtained. The estimation of divergences of class pairs is unstable using inadequate labeled data. To take advantage of unlabeled data for subspace selection, semi-supervised MGMD (SSMGMD) is proposed using graph Laplacian as normalization. Quasi-Newton method is adopted to solve the optimization problem. Experiments on synthetic data and real image data show the validity of SSMGMD.</description><subject>Covariance matrix</subject><subject>Educational institutions</subject><subject>Laplace equations</subject><subject>Manifolds</subject><subject>Optimization</subject><subject>Symmetric matrices</subject><subject>Training</subject><isbn>9781612841809</isbn><isbn>1612841805</isbn><isbn>9781612841816</isbn><isbn>1612841813</isbn><isbn>1612841791</isbn><isbn>9781612841793</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2011</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><sourceid>RIE</sourceid><recordid>eNpVkL1OwzAUhY0QEqjkARCLXyDBjhPbd0SFUtRIDC1zZSfXlSF_spNKvD2R6MIZzqdvOcMh5IGzjHMGT5v97iXLGeeZZBwUz69IAkpzyXNd8IXX_5zBLUli_GJLpASh9B353GPn0ziPGM4-YkNPOHQ4BV_TDk1PB0d3c9taU3-nFXrbYqCNP2M4YV9jpG4INM42jqZGGrHFevJDf09unGkjJheuyGHzelhv0-rj7X39XKUe2JQWTIOwrlAOHHMsl8xqa4QupEPrykIKUEzlUAKU3NilSmWsUY1uTC4FihV5_Jv1iHgcg-9M-DlevhC_5_1Smw</recordid><startdate>201107</startdate><enddate>201107</enddate><creator>Si-Bao Chen</creator><creator>Hai-Xian Wang</creator><creator>Xing-Yi Zhang</creator><creator>Bin Luo</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope></search><sort><creationdate>201107</creationdate><title>Semi-supervised geometric mean of Kullback-Leibler divergences for subspace selection</title><author>Si-Bao Chen ; Hai-Xian Wang ; Xing-Yi Zhang ; Bin Luo</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i90t-40893bf47f9f0f0260b8ba3846febf546397072959951ab95157aba7d8da263e3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2011</creationdate><topic>Covariance matrix</topic><topic>Educational institutions</topic><topic>Laplace equations</topic><topic>Manifolds</topic><topic>Optimization</topic><topic>Symmetric matrices</topic><topic>Training</topic><toplevel>online_resources</toplevel><creatorcontrib>Si-Bao Chen</creatorcontrib><creatorcontrib>Hai-Xian Wang</creatorcontrib><creatorcontrib>Xing-Yi Zhang</creatorcontrib><creatorcontrib>Bin Luo</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan (POP) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Si-Bao Chen</au><au>Hai-Xian Wang</au><au>Xing-Yi Zhang</au><au>Bin Luo</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Semi-supervised geometric mean of Kullback-Leibler divergences for subspace selection</atitle><btitle>2011 Eighth International Conference on Fuzzy Systems and Knowledge Discovery (FSKD)</btitle><stitle>FSKD</stitle><date>2011-07</date><risdate>2011</risdate><volume>2</volume><spage>1232</spage><epage>1235</epage><pages>1232-1235</pages><isbn>9781612841809</isbn><isbn>1612841805</isbn><eisbn>9781612841816</eisbn><eisbn>1612841813</eisbn><eisbn>1612841791</eisbn><eisbn>9781612841793</eisbn><abstract>Subspace selection is widely adopted in many areas of pattern recognition. A recent result, named maximizing the geometric mean of Kullback-Leibler (KL) divergences of class pairs (MGMD), is a successful method for subspace selection, which can significantly reduce the class separation problem. However, in many applications, labeled data are very limited while unlabeled data can be easily obtained. The estimation of divergences of class pairs is unstable using inadequate labeled data. To take advantage of unlabeled data for subspace selection, semi-supervised MGMD (SSMGMD) is proposed using graph Laplacian as normalization. Quasi-Newton method is adopted to solve the optimization problem. Experiments on synthetic data and real image data show the validity of SSMGMD.</abstract><pub>IEEE</pub><doi>10.1109/FSKD.2011.6019712</doi><tpages>4</tpages></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISBN: 9781612841809 |
| ispartof | 2011 Eighth International Conference on Fuzzy Systems and Knowledge Discovery (FSKD), 2011, Vol.2, p.1232-1235 |
| issn | |
| language | eng |
| recordid | cdi_ieee_primary_6019712 |
| source | IEEE Electronic Library (IEL) Conference Proceedings |
| subjects | Covariance matrix Educational institutions Laplace equations Manifolds Optimization Symmetric matrices Training |
| title | Semi-supervised geometric mean of Kullback-Leibler divergences for subspace selection |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-28T04%3A29%3A12IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Semi-supervised%20geometric%20mean%20of%20Kullback-Leibler%20divergences%20for%20subspace%20selection&rft.btitle=2011%20Eighth%20International%20Conference%20on%20Fuzzy%20Systems%20and%20Knowledge%20Discovery%20(FSKD)&rft.au=Si-Bao%20Chen&rft.date=2011-07&rft.volume=2&rft.spage=1232&rft.epage=1235&rft.pages=1232-1235&rft.isbn=9781612841809&rft.isbn_list=1612841805&rft_id=info:doi/10.1109/FSKD.2011.6019712&rft_dat=%3Cieee_6IE%3E6019712%3C/ieee_6IE%3E%3Curl%3E%3C/url%3E&rft.eisbn=9781612841816&rft.eisbn_list=1612841813&rft.eisbn_list=1612841791&rft.eisbn_list=9781612841793&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=6019712&rfr_iscdi=true |