TopRank+: A Refinement of TopRank Algorithm

TopRank+: A Refinement of TopRank Algorithm

Online learning to rank is a core problem in machine learning. In Lattimore et al. (2018), a novel online learning algorithm was proposed based on topological sorting. In the paper they provided a set of self-normalized inequalities (a) in the algorithm as a criterion in iterations and (b) to provid...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: de la Pena, Victor, Zou, Haolin
Format: Artikel
Sprache:eng
Schlagworte:
Computer Science - Learning
Statistics - Machine Learning
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator de la Pena, Victor
Zou, Haolin
description Online learning to rank is a core problem in machine learning. In Lattimore et al. (2018), a novel online learning algorithm was proposed based on topological sorting. In the paper they provided a set of self-normalized inequalities (a) in the algorithm as a criterion in iterations and (b) to provide an upper bound for cumulative regret, which is a measure of algorithm performance. In this work, we utilized method of mixtures and asymptotic expansions of certain implicit function to provide a tighter, iterated-log-like boundary for the inequalities, and as a consequence improve both the algorithm itself as well as its performance estimation.
doi_str_mv 10.48550/arxiv.2001.07617
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2001_07617</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2001_07617</sourcerecordid><originalsourceid>FETCH-LOGICAL-a677-effbbad193c235f77702c6533b5b0fe37f93dd29c231930b7e5436e2992369c33</originalsourceid><addsrcrecordid>eNotjr0KwjAYRbM4iPoATmaX1jSfyWfcSvEPCoJ0L0mbaNG2Eovo2_s73eFcDoeQccTC-UIINtP-Ud1DzlgUMpQR9sk0a68H3ZynSxrTg3VVY2vbdLR19E9ofDm2vupO9ZD0nL7c7Oi_A5KtV1myDdL9ZpfEaaAlYmCdM0aXkYKCg3CIyHghBYARhjkL6BSUJVdv-v4wg1bMQVquFAepCoABmfy039r86qta-2f-qc6_1fACF5Q6oA</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>TopRank+: A Refinement of TopRank Algorithm</title><source>arXiv.org</source><creator>de la Pena, Victor ; Zou, Haolin</creator><creatorcontrib>de la Pena, Victor ; Zou, Haolin</creatorcontrib><description>Online learning to rank is a core problem in machine learning. In Lattimore et al. (2018), a novel online learning algorithm was proposed based on topological sorting. In the paper they provided a set of self-normalized inequalities (a) in the algorithm as a criterion in iterations and (b) to provide an upper bound for cumulative regret, which is a measure of algorithm performance. In this work, we utilized method of mixtures and asymptotic expansions of certain implicit function to provide a tighter, iterated-log-like boundary for the inequalities, and as a consequence improve both the algorithm itself as well as its performance estimation.</description><identifier>DOI: 10.48550/arxiv.2001.07617</identifier><language>eng</language><subject>Computer Science - Learning ; Statistics - Machine Learning</subject><creationdate>2020-01</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2001.07617$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2001.07617$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>de la Pena, Victor</creatorcontrib><creatorcontrib>Zou, Haolin</creatorcontrib><title>TopRank+: A Refinement of TopRank Algorithm</title><description>Online learning to rank is a core problem in machine learning. In Lattimore et al. (2018), a novel online learning algorithm was proposed based on topological sorting. In the paper they provided a set of self-normalized inequalities (a) in the algorithm as a criterion in iterations and (b) to provide an upper bound for cumulative regret, which is a measure of algorithm performance. In this work, we utilized method of mixtures and asymptotic expansions of certain implicit function to provide a tighter, iterated-log-like boundary for the inequalities, and as a consequence improve both the algorithm itself as well as its performance estimation.</description><subject>Computer Science - Learning</subject><subject>Statistics - Machine Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotjr0KwjAYRbM4iPoATmaX1jSfyWfcSvEPCoJ0L0mbaNG2Eovo2_s73eFcDoeQccTC-UIINtP-Ud1DzlgUMpQR9sk0a68H3ZynSxrTg3VVY2vbdLR19E9ofDm2vupO9ZD0nL7c7Oi_A5KtV1myDdL9ZpfEaaAlYmCdM0aXkYKCg3CIyHghBYARhjkL6BSUJVdv-v4wg1bMQVquFAepCoABmfy039r86qta-2f-qc6_1fACF5Q6oA</recordid><startdate>20200121</startdate><enddate>20200121</enddate><creator>de la Pena, Victor</creator><creator>Zou, Haolin</creator><scope>AKY</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20200121</creationdate><title>TopRank+: A Refinement of TopRank Algorithm</title><author>de la Pena, Victor ; Zou, Haolin</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a677-effbbad193c235f77702c6533b5b0fe37f93dd29c231930b7e5436e2992369c33</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Computer Science - Learning</topic><topic>Statistics - Machine Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>de la Pena, Victor</creatorcontrib><creatorcontrib>Zou, Haolin</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>de la Pena, Victor</au><au>Zou, Haolin</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>TopRank+: A Refinement of TopRank Algorithm</atitle><date>2020-01-21</date><risdate>2020</risdate><abstract>Online learning to rank is a core problem in machine learning. In Lattimore et al. (2018), a novel online learning algorithm was proposed based on topological sorting. In the paper they provided a set of self-normalized inequalities (a) in the algorithm as a criterion in iterations and (b) to provide an upper bound for cumulative regret, which is a measure of algorithm performance. In this work, we utilized method of mixtures and asymptotic expansions of certain implicit function to provide a tighter, iterated-log-like boundary for the inequalities, and as a consequence improve both the algorithm itself as well as its performance estimation.</abstract><doi>10.48550/arxiv.2001.07617</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2001.07617
ispartof
issn
language eng
recordid cdi_arxiv_primary_2001_07617
source arXiv.org
subjects Computer Science - Learning
Statistics - Machine Learning
title TopRank+: A Refinement of TopRank Algorithm
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-25T14%3A20%3A16IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=TopRank+:%20A%20Refinement%20of%20TopRank%20Algorithm&rft.au=de%20la%20Pena,%20Victor&rft.date=2020-01-21&rft_id=info:doi/10.48550/arxiv.2001.07617&rft_dat=%3Carxiv_GOX%3E2001_07617%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true