Loading…

Local variational Probabilistic Minimax Active Learning

In the last decade, many excellent active learning methods have been proposed whose algorithms generally deliver an acceptable performance. However, many of these methods rely on measures other than the risk of the classifier. Recently, researchers have proposed Probabilistic Minimax Active Learning...

Full description

Saved in:
Bibliographic Details
Published in:Expert systems with applications 2023-01, Vol.211, p.118538, Article 118538
Main Author: Ghafarian, Seyed Hossein
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c230t-474747fca45a7589d8d9afbd85ac81cdb93469b5da9ea3728d50fcf5c5cd41683
cites cdi_FETCH-LOGICAL-c230t-474747fca45a7589d8d9afbd85ac81cdb93469b5da9ea3728d50fcf5c5cd41683
container_end_page
container_issue
container_start_page 118538
container_title Expert systems with applications
container_volume 211
creator Ghafarian, Seyed Hossein
description In the last decade, many excellent active learning methods have been proposed whose algorithms generally deliver an acceptable performance. However, many of these methods rely on measures other than the risk of the classifier. Recently, researchers have proposed Probabilistic Minimax Active Learning (PMAL). PMAL is optimal because, after adding the result of a query, it minimizes the upper bound of the risk of the classifier. Unfortunately, the exact computation of PMAL’s objective function is intractable for likelihood functions suitable for classification problems, such as logistic regression. Employing a variational approach, the present study approximates the PMAL objective when logistic regression is the likelihood function. Experiments show that the proposed algorithm effectively asks queries, which results in superior performance to that of the state-of-the-art.  •Logistic regression which is a workhorse for statistical learning, is used to estimate Probabilistic Minimax Active Learning (PMAL) objective function.•A Local variational approximation of PMAL with logistic regression has been proposed.•An active learning algorithm based on the local variational approximation of logistic PMAL objective shows superior results.
doi_str_mv 10.1016/j.eswa.2022.118538
format article
fullrecord <record><control><sourceid>elsevier_cross</sourceid><recordid>TN_cdi_crossref_primary_10_1016_j_eswa_2022_118538</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S0957417422016128</els_id><sourcerecordid>S0957417422016128</sourcerecordid><originalsourceid>FETCH-LOGICAL-c230t-474747fca45a7589d8d9afbd85ac81cdb93469b5da9ea3728d50fcf5c5cd41683</originalsourceid><addsrcrecordid>eNp9j8FKxDAURYMoOI7-gKv-QGvSNE0CboZBHaGiC12H15dU3jC2kpRR_96Wupa7eHdzHvcwdi14Ibiob_ZFSF9QlLwsCyGMkuaErYTRMq-1ladsxa3SeSV0dc4uUtpzLjTnesV0MyAcsiNEgpGGfuovcWihpQOlkTB7op4-4Dvb4EjHkDUBYk_9-yU76-CQwtXfXbO3-7vX7S5vnh8et5smx1LyMa_0nA6hUqCVsd54C13rjQI0An1rZVXbVnmwAaQujVe8w06hQl-J2sg1K5e_GIeUYujcZ5z2xB8nuJvV3d7N6m5Wd4v6BN0uUJiWHSlEl5BCj8FTDDg6P9B_-C_fK2K6</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Local variational Probabilistic Minimax Active Learning</title><source>ScienceDirect Journals</source><creator>Ghafarian, Seyed Hossein</creator><creatorcontrib>Ghafarian, Seyed Hossein</creatorcontrib><description>In the last decade, many excellent active learning methods have been proposed whose algorithms generally deliver an acceptable performance. However, many of these methods rely on measures other than the risk of the classifier. Recently, researchers have proposed Probabilistic Minimax Active Learning (PMAL). PMAL is optimal because, after adding the result of a query, it minimizes the upper bound of the risk of the classifier. Unfortunately, the exact computation of PMAL’s objective function is intractable for likelihood functions suitable for classification problems, such as logistic regression. Employing a variational approach, the present study approximates the PMAL objective when logistic regression is the likelihood function. Experiments show that the proposed algorithm effectively asks queries, which results in superior performance to that of the state-of-the-art.  •Logistic regression which is a workhorse for statistical learning, is used to estimate Probabilistic Minimax Active Learning (PMAL) objective function.•A Local variational approximation of PMAL with logistic regression has been proposed.•An active learning algorithm based on the local variational approximation of logistic PMAL objective shows superior results.</description><identifier>ISSN: 0957-4174</identifier><identifier>EISSN: 1873-6793</identifier><identifier>DOI: 10.1016/j.eswa.2022.118538</identifier><language>eng</language><publisher>Elsevier Ltd</publisher><subject>Local variational ; Logistic regression ; Probabilistic Minimax Active Learning</subject><ispartof>Expert systems with applications, 2023-01, Vol.211, p.118538, Article 118538</ispartof><rights>2022 Elsevier Ltd</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c230t-474747fca45a7589d8d9afbd85ac81cdb93469b5da9ea3728d50fcf5c5cd41683</citedby><cites>FETCH-LOGICAL-c230t-474747fca45a7589d8d9afbd85ac81cdb93469b5da9ea3728d50fcf5c5cd41683</cites><orcidid>0000-0001-6557-5129</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27924,27925</link.rule.ids></links><search><creatorcontrib>Ghafarian, Seyed Hossein</creatorcontrib><title>Local variational Probabilistic Minimax Active Learning</title><title>Expert systems with applications</title><description>In the last decade, many excellent active learning methods have been proposed whose algorithms generally deliver an acceptable performance. However, many of these methods rely on measures other than the risk of the classifier. Recently, researchers have proposed Probabilistic Minimax Active Learning (PMAL). PMAL is optimal because, after adding the result of a query, it minimizes the upper bound of the risk of the classifier. Unfortunately, the exact computation of PMAL’s objective function is intractable for likelihood functions suitable for classification problems, such as logistic regression. Employing a variational approach, the present study approximates the PMAL objective when logistic regression is the likelihood function. Experiments show that the proposed algorithm effectively asks queries, which results in superior performance to that of the state-of-the-art.  •Logistic regression which is a workhorse for statistical learning, is used to estimate Probabilistic Minimax Active Learning (PMAL) objective function.•A Local variational approximation of PMAL with logistic regression has been proposed.•An active learning algorithm based on the local variational approximation of logistic PMAL objective shows superior results.</description><subject>Local variational</subject><subject>Logistic regression</subject><subject>Probabilistic Minimax Active Learning</subject><issn>0957-4174</issn><issn>1873-6793</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><recordid>eNp9j8FKxDAURYMoOI7-gKv-QGvSNE0CboZBHaGiC12H15dU3jC2kpRR_96Wupa7eHdzHvcwdi14Ibiob_ZFSF9QlLwsCyGMkuaErYTRMq-1ladsxa3SeSV0dc4uUtpzLjTnesV0MyAcsiNEgpGGfuovcWihpQOlkTB7op4-4Dvb4EjHkDUBYk_9-yU76-CQwtXfXbO3-7vX7S5vnh8et5smx1LyMa_0nA6hUqCVsd54C13rjQI0An1rZVXbVnmwAaQujVe8w06hQl-J2sg1K5e_GIeUYujcZ5z2xB8nuJvV3d7N6m5Wd4v6BN0uUJiWHSlEl5BCj8FTDDg6P9B_-C_fK2K6</recordid><startdate>202301</startdate><enddate>202301</enddate><creator>Ghafarian, Seyed Hossein</creator><general>Elsevier Ltd</general><scope>AAYXX</scope><scope>CITATION</scope><orcidid>https://orcid.org/0000-0001-6557-5129</orcidid></search><sort><creationdate>202301</creationdate><title>Local variational Probabilistic Minimax Active Learning</title><author>Ghafarian, Seyed Hossein</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c230t-474747fca45a7589d8d9afbd85ac81cdb93469b5da9ea3728d50fcf5c5cd41683</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Local variational</topic><topic>Logistic regression</topic><topic>Probabilistic Minimax Active Learning</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Ghafarian, Seyed Hossein</creatorcontrib><collection>CrossRef</collection><jtitle>Expert systems with applications</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Ghafarian, Seyed Hossein</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Local variational Probabilistic Minimax Active Learning</atitle><jtitle>Expert systems with applications</jtitle><date>2023-01</date><risdate>2023</risdate><volume>211</volume><spage>118538</spage><pages>118538-</pages><artnum>118538</artnum><issn>0957-4174</issn><eissn>1873-6793</eissn><abstract>In the last decade, many excellent active learning methods have been proposed whose algorithms generally deliver an acceptable performance. However, many of these methods rely on measures other than the risk of the classifier. Recently, researchers have proposed Probabilistic Minimax Active Learning (PMAL). PMAL is optimal because, after adding the result of a query, it minimizes the upper bound of the risk of the classifier. Unfortunately, the exact computation of PMAL’s objective function is intractable for likelihood functions suitable for classification problems, such as logistic regression. Employing a variational approach, the present study approximates the PMAL objective when logistic regression is the likelihood function. Experiments show that the proposed algorithm effectively asks queries, which results in superior performance to that of the state-of-the-art.  •Logistic regression which is a workhorse for statistical learning, is used to estimate Probabilistic Minimax Active Learning (PMAL) objective function.•A Local variational approximation of PMAL with logistic regression has been proposed.•An active learning algorithm based on the local variational approximation of logistic PMAL objective shows superior results.</abstract><pub>Elsevier Ltd</pub><doi>10.1016/j.eswa.2022.118538</doi><orcidid>https://orcid.org/0000-0001-6557-5129</orcidid></addata></record>
fulltext fulltext
identifier ISSN: 0957-4174
ispartof Expert systems with applications, 2023-01, Vol.211, p.118538, Article 118538
issn 0957-4174
1873-6793
language eng
recordid cdi_crossref_primary_10_1016_j_eswa_2022_118538
source ScienceDirect Journals
subjects Local variational
Logistic regression
Probabilistic Minimax Active Learning
title Local variational Probabilistic Minimax Active Learning
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T15%3A09%3A34IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-elsevier_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Local%20variational%20Probabilistic%20Minimax%20Active%20Learning&rft.jtitle=Expert%20systems%20with%20applications&rft.au=Ghafarian,%20Seyed%20Hossein&rft.date=2023-01&rft.volume=211&rft.spage=118538&rft.pages=118538-&rft.artnum=118538&rft.issn=0957-4174&rft.eissn=1873-6793&rft_id=info:doi/10.1016/j.eswa.2022.118538&rft_dat=%3Celsevier_cross%3ES0957417422016128%3C/elsevier_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c230t-474747fca45a7589d8d9afbd85ac81cdb93469b5da9ea3728d50fcf5c5cd41683%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true