A Probabilistic Active Learning Algorithm Based on Fisher Information Ratio
The task of labeling samples is demanding and expensive. Active learning aims to generate the smallest possible training data set that results in a classifier with high performance in the test phase. It usually consists of two steps of selecting a set of queries and requesting their labels. Among th...
Gespeichert in:
| Veröffentlicht in: | IEEE transactions on pattern analysis and machine intelligence 2018-08, Vol.40 (8), p.2023-2029 |
|---|---|
| Hauptverfasser: | , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 2029 |
|---|---|
| container_issue | 8 |
| container_start_page | 2023 |
| container_title | IEEE transactions on pattern analysis and machine intelligence |
| container_volume | 40 |
| creator | Sourati, Jamshid Akcakaya, Murat Erdogmus, Deniz Leen, Todd K. Dy, Jennifer G. |
| description | The task of labeling samples is demanding and expensive. Active learning aims to generate the smallest possible training data set that results in a classifier with high performance in the test phase. It usually consists of two steps of selecting a set of queries and requesting their labels. Among the suggested objectives to score the query sets, information theoretic measures have become very popular. Yet among them, those based on Fisher information (FI) have the advantage of considering the diversity among the queries and tractable computations. In this work, we provide a practical algorithm based on Fisher information ratio to obtain query distribution for a general framework where, in contrast to the previous FI-based querying methods, we make no assumptions over the test distribution. The empirical results on synthetic and real-world data sets indicate that this algorithm gives competitive results. |
| doi_str_mv | 10.1109/TPAMI.2017.2743707 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_RIE</sourceid><recordid>TN_cdi_crossref_primary_10_1109_TPAMI_2017_2743707</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>8016395</ieee_id><sourcerecordid>1988258157</sourcerecordid><originalsourceid>FETCH-LOGICAL-c395t-ed9da20dd1084595bb0bf3379771a127b713a7176e56441e44ba77cca3697f913</originalsourceid><addsrcrecordid>eNpdkUtLAzEQx4MoWh9fQEEWvHjZmsmjkxxX8VGsKKLnkN3Naso-NNkKfnu3tvbgaSDzmz8zvxByDHQMQPXFy1P2MB0zCjhmKDhS3CIj0FynXHK9TUYUJixViqk9sh_jnFIQkvJdsseUkgqVGJH7LHkKXW5zX_vY-yLJit5_uWTmbGh9-5Zk9VsXfP_eJJc2ujLp2uTGx3cXkmlbdaGxvR-enpflkOxUto7uaF0PyOvN9cvVXTp7vJ1eZbO04Fr2qSt1aRktS6BKSC3znOYV56gRwQLDHIFbBJw4ORECnBC5RSwKyycaKw38gJyvcj9C97lwsTeNj4Wra9u6bhEN6OFmqUDigJ79Q-fdIrTDdoYBCqE4Y8tAtqKK0MUYXGU-gm9s-DZAzVK1-VVtlqrNWvUwdLqOXuSNKzcjf24H4GQFeOfcpq2GTxk08B-XTICE</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2174483221</pqid></control><display><type>article</type><title>A Probabilistic Active Learning Algorithm Based on Fisher Information Ratio</title><source>IEEE Electronic Library (IEL)</source><creator>Sourati, Jamshid ; Akcakaya, Murat ; Erdogmus, Deniz ; Leen, Todd K. ; Dy, Jennifer G.</creator><creatorcontrib>Sourati, Jamshid ; Akcakaya, Murat ; Erdogmus, Deniz ; Leen, Todd K. ; Dy, Jennifer G.</creatorcontrib><description>The task of labeling samples is demanding and expensive. Active learning aims to generate the smallest possible training data set that results in a classifier with high performance in the test phase. It usually consists of two steps of selecting a set of queries and requesting their labels. Among the suggested objectives to score the query sets, information theoretic measures have become very popular. Yet among them, those based on Fisher information (FI) have the advantage of considering the diversity among the queries and tractable computations. In this work, we provide a practical algorithm based on Fisher information ratio to obtain query distribution for a general framework where, in contrast to the previous FI-based querying methods, we make no assumptions over the test distribution. The empirical results on synthetic and real-world data sets indicate that this algorithm gives competitive results.</description><identifier>ISSN: 0162-8828</identifier><identifier>EISSN: 1939-3539</identifier><identifier>EISSN: 2160-9292</identifier><identifier>DOI: 10.1109/TPAMI.2017.2743707</identifier><identifier>PMID: 28858784</identifier><identifier>CODEN: ITPIDJ</identifier><language>eng</language><publisher>United States: IEEE</publisher><subject>Active learning ; Algorithms ; Approximation algorithms ; Computational complexity ; Computer Simulation ; Databases, Factual - statistics & numerical data ; discriminative classification ; Finite impulse response filters ; Fisher information ; Humans ; Information theory ; Labels ; Machine learning ; Models, Statistical ; Monte Carlo Method ; Optimization ; Probabilistic logic ; probabilistic querying ; Proposals ; Queries ; Training</subject><ispartof>IEEE transactions on pattern analysis and machine intelligence, 2018-08, Vol.40 (8), p.2023-2029</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2018</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c395t-ed9da20dd1084595bb0bf3379771a127b713a7176e56441e44ba77cca3697f913</citedby><cites>FETCH-LOGICAL-c395t-ed9da20dd1084595bb0bf3379771a127b713a7176e56441e44ba77cca3697f913</cites><orcidid>0000-0003-1853-7271</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/8016395$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>314,780,784,796,27924,27925,54758</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/8016395$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/28858784$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Sourati, Jamshid</creatorcontrib><creatorcontrib>Akcakaya, Murat</creatorcontrib><creatorcontrib>Erdogmus, Deniz</creatorcontrib><creatorcontrib>Leen, Todd K.</creatorcontrib><creatorcontrib>Dy, Jennifer G.</creatorcontrib><title>A Probabilistic Active Learning Algorithm Based on Fisher Information Ratio</title><title>IEEE transactions on pattern analysis and machine intelligence</title><addtitle>TPAMI</addtitle><addtitle>IEEE Trans Pattern Anal Mach Intell</addtitle><description>The task of labeling samples is demanding and expensive. Active learning aims to generate the smallest possible training data set that results in a classifier with high performance in the test phase. It usually consists of two steps of selecting a set of queries and requesting their labels. Among the suggested objectives to score the query sets, information theoretic measures have become very popular. Yet among them, those based on Fisher information (FI) have the advantage of considering the diversity among the queries and tractable computations. In this work, we provide a practical algorithm based on Fisher information ratio to obtain query distribution for a general framework where, in contrast to the previous FI-based querying methods, we make no assumptions over the test distribution. The empirical results on synthetic and real-world data sets indicate that this algorithm gives competitive results.</description><subject>Active learning</subject><subject>Algorithms</subject><subject>Approximation algorithms</subject><subject>Computational complexity</subject><subject>Computer Simulation</subject><subject>Databases, Factual - statistics & numerical data</subject><subject>discriminative classification</subject><subject>Finite impulse response filters</subject><subject>Fisher information</subject><subject>Humans</subject><subject>Information theory</subject><subject>Labels</subject><subject>Machine learning</subject><subject>Models, Statistical</subject><subject>Monte Carlo Method</subject><subject>Optimization</subject><subject>Probabilistic logic</subject><subject>probabilistic querying</subject><subject>Proposals</subject><subject>Queries</subject><subject>Training</subject><issn>0162-8828</issn><issn>1939-3539</issn><issn>2160-9292</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2018</creationdate><recordtype>article</recordtype><sourceid>RIE</sourceid><sourceid>EIF</sourceid><recordid>eNpdkUtLAzEQx4MoWh9fQEEWvHjZmsmjkxxX8VGsKKLnkN3Naso-NNkKfnu3tvbgaSDzmz8zvxByDHQMQPXFy1P2MB0zCjhmKDhS3CIj0FynXHK9TUYUJixViqk9sh_jnFIQkvJdsseUkgqVGJH7LHkKXW5zX_vY-yLJit5_uWTmbGh9-5Zk9VsXfP_eJJc2ujLp2uTGx3cXkmlbdaGxvR-enpflkOxUto7uaF0PyOvN9cvVXTp7vJ1eZbO04Fr2qSt1aRktS6BKSC3znOYV56gRwQLDHIFbBJw4ORECnBC5RSwKyycaKw38gJyvcj9C97lwsTeNj4Wra9u6bhEN6OFmqUDigJ79Q-fdIrTDdoYBCqE4Y8tAtqKK0MUYXGU-gm9s-DZAzVK1-VVtlqrNWvUwdLqOXuSNKzcjf24H4GQFeOfcpq2GTxk08B-XTICE</recordid><startdate>20180801</startdate><enddate>20180801</enddate><creator>Sourati, Jamshid</creator><creator>Akcakaya, Murat</creator><creator>Erdogmus, Deniz</creator><creator>Leen, Todd K.</creator><creator>Dy, Jennifer G.</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0003-1853-7271</orcidid></search><sort><creationdate>20180801</creationdate><title>A Probabilistic Active Learning Algorithm Based on Fisher Information Ratio</title><author>Sourati, Jamshid ; Akcakaya, Murat ; Erdogmus, Deniz ; Leen, Todd K. ; Dy, Jennifer G.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c395t-ed9da20dd1084595bb0bf3379771a127b713a7176e56441e44ba77cca3697f913</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Active learning</topic><topic>Algorithms</topic><topic>Approximation algorithms</topic><topic>Computational complexity</topic><topic>Computer Simulation</topic><topic>Databases, Factual - statistics & numerical data</topic><topic>discriminative classification</topic><topic>Finite impulse response filters</topic><topic>Fisher information</topic><topic>Humans</topic><topic>Information theory</topic><topic>Labels</topic><topic>Machine learning</topic><topic>Models, Statistical</topic><topic>Monte Carlo Method</topic><topic>Optimization</topic><topic>Probabilistic logic</topic><topic>probabilistic querying</topic><topic>Proposals</topic><topic>Queries</topic><topic>Training</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Sourati, Jamshid</creatorcontrib><creatorcontrib>Akcakaya, Murat</creatorcontrib><creatorcontrib>Erdogmus, Deniz</creatorcontrib><creatorcontrib>Leen, Todd K.</creatorcontrib><creatorcontrib>Dy, Jennifer G.</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>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics & Communications Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest Computer Science 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>MEDLINE - Academic</collection><jtitle>IEEE transactions on pattern analysis and machine intelligence</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Sourati, Jamshid</au><au>Akcakaya, Murat</au><au>Erdogmus, Deniz</au><au>Leen, Todd K.</au><au>Dy, Jennifer G.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A Probabilistic Active Learning Algorithm Based on Fisher Information Ratio</atitle><jtitle>IEEE transactions on pattern analysis and machine intelligence</jtitle><stitle>TPAMI</stitle><addtitle>IEEE Trans Pattern Anal Mach Intell</addtitle><date>2018-08-01</date><risdate>2018</risdate><volume>40</volume><issue>8</issue><spage>2023</spage><epage>2029</epage><pages>2023-2029</pages><issn>0162-8828</issn><eissn>1939-3539</eissn><eissn>2160-9292</eissn><coden>ITPIDJ</coden><abstract>The task of labeling samples is demanding and expensive. Active learning aims to generate the smallest possible training data set that results in a classifier with high performance in the test phase. It usually consists of two steps of selecting a set of queries and requesting their labels. Among the suggested objectives to score the query sets, information theoretic measures have become very popular. Yet among them, those based on Fisher information (FI) have the advantage of considering the diversity among the queries and tractable computations. In this work, we provide a practical algorithm based on Fisher information ratio to obtain query distribution for a general framework where, in contrast to the previous FI-based querying methods, we make no assumptions over the test distribution. The empirical results on synthetic and real-world data sets indicate that this algorithm gives competitive results.</abstract><cop>United States</cop><pub>IEEE</pub><pmid>28858784</pmid><doi>10.1109/TPAMI.2017.2743707</doi><tpages>7</tpages><orcidid>https://orcid.org/0000-0003-1853-7271</orcidid><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 0162-8828 |
| ispartof | IEEE transactions on pattern analysis and machine intelligence, 2018-08, Vol.40 (8), p.2023-2029 |
| issn | 0162-8828 1939-3539 2160-9292 |
| language | eng |
| recordid | cdi_crossref_primary_10_1109_TPAMI_2017_2743707 |
| source | IEEE Electronic Library (IEL) |
| subjects | Active learning Algorithms Approximation algorithms Computational complexity Computer Simulation Databases, Factual - statistics & numerical data discriminative classification Finite impulse response filters Fisher information Humans Information theory Labels Machine learning Models, Statistical Monte Carlo Method Optimization Probabilistic logic probabilistic querying Proposals Queries Training |
| title | A Probabilistic Active Learning Algorithm Based on Fisher Information Ratio |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-22T10%3A05%3A15IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_RIE&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20Probabilistic%20Active%20Learning%20Algorithm%20Based%20on%20Fisher%20Information%20Ratio&rft.jtitle=IEEE%20transactions%20on%20pattern%20analysis%20and%20machine%20intelligence&rft.au=Sourati,%20Jamshid&rft.date=2018-08-01&rft.volume=40&rft.issue=8&rft.spage=2023&rft.epage=2029&rft.pages=2023-2029&rft.issn=0162-8828&rft.eissn=1939-3539&rft.coden=ITPIDJ&rft_id=info:doi/10.1109/TPAMI.2017.2743707&rft_dat=%3Cproquest_RIE%3E1988258157%3C/proquest_RIE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2174483221&rft_id=info:pmid/28858784&rft_ieee_id=8016395&rfr_iscdi=true |