A KL-LUCB Bandit Algorithm for Large-Scale Crowdsourcing

This paper focuses on best-arm identification in multi-armed bandits with bounded rewards. We develop an algorithm that is a fusion of lil-UCB and KL-LUCB, offering the best qualities of the two algorithms in one method. This is achieved by proving a novel anytime confidence bound for the mean of bo...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2017-09
Main Authors: Mankoff, Bob, Nowak, Robert, Ervin Tanczos
Format: Article
Language:English
Subjects:
Algorithms
Crowdsourcing
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Mankoff, Bob
Nowak, Robert
Ervin Tanczos
description This paper focuses on best-arm identification in multi-armed bandits with bounded rewards. We develop an algorithm that is a fusion of lil-UCB and KL-LUCB, offering the best qualities of the two algorithms in one method. This is achieved by proving a novel anytime confidence bound for the mean of bounded distributions, which is the analogue of the LIL-type bounds recently developed for sub-Gaussian distributions. We corroborate our theoretical results with numerical experiments based on the New Yorker Cartoon Caption Contest.
format article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2076865905</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2076865905</sourcerecordid><originalsourceid>FETCH-proquest_journals_20768659053</originalsourceid><addsrcrecordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mSwcFTw9tH1CXV2UnBKzEvJLFFwzEnPL8osychVSMsvUvBJLEpP1Q1OTsxJVXAuyi9PKc4vLUrOzEvnYWBNS8wpTuWF0twMym6uIc4eugVF-YWlqcUl8VlAlXlAqXgjA3MzCzNTSwNTY-JUAQB0HzSx</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2076865905</pqid></control><display><type>article</type><title>A KL-LUCB Bandit Algorithm for Large-Scale Crowdsourcing</title><source>ProQuest - Publicly Available Content Database</source><creator>Mankoff, Bob ; Nowak, Robert ; Ervin Tanczos</creator><creatorcontrib>Mankoff, Bob ; Nowak, Robert ; Ervin Tanczos</creatorcontrib><description>This paper focuses on best-arm identification in multi-armed bandits with bounded rewards. We develop an algorithm that is a fusion of lil-UCB and KL-LUCB, offering the best qualities of the two algorithms in one method. This is achieved by proving a novel anytime confidence bound for the mean of bounded distributions, which is the analogue of the LIL-type bounds recently developed for sub-Gaussian distributions. We corroborate our theoretical results with numerical experiments based on the New Yorker Cartoon Caption Contest.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Algorithms ; Crowdsourcing</subject><ispartof>arXiv.org, 2017-09</ispartof><rights>2017. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/2076865905?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>780,784,25753,37012,44590</link.rule.ids></links><search><creatorcontrib>Mankoff, Bob</creatorcontrib><creatorcontrib>Nowak, Robert</creatorcontrib><creatorcontrib>Ervin Tanczos</creatorcontrib><title>A KL-LUCB Bandit Algorithm for Large-Scale Crowdsourcing</title><title>arXiv.org</title><description>This paper focuses on best-arm identification in multi-armed bandits with bounded rewards. We develop an algorithm that is a fusion of lil-UCB and KL-LUCB, offering the best qualities of the two algorithms in one method. This is achieved by proving a novel anytime confidence bound for the mean of bounded distributions, which is the analogue of the LIL-type bounds recently developed for sub-Gaussian distributions. We corroborate our theoretical results with numerical experiments based on the New Yorker Cartoon Caption Contest.</description><subject>Algorithms</subject><subject>Crowdsourcing</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2017</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mSwcFTw9tH1CXV2UnBKzEvJLFFwzEnPL8osychVSMsvUvBJLEpP1Q1OTsxJVXAuyi9PKc4vLUrOzEvnYWBNS8wpTuWF0twMym6uIc4eugVF-YWlqcUl8VlAlXlAqXgjA3MzCzNTSwNTY-JUAQB0HzSx</recordid><startdate>20170911</startdate><enddate>20170911</enddate><creator>Mankoff, Bob</creator><creator>Nowak, Robert</creator><creator>Ervin Tanczos</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20170911</creationdate><title>A KL-LUCB Bandit Algorithm for Large-Scale Crowdsourcing</title><author>Mankoff, Bob ; Nowak, Robert ; Ervin Tanczos</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_20768659053</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2017</creationdate><topic>Algorithms</topic><topic>Crowdsourcing</topic><toplevel>online_resources</toplevel><creatorcontrib>Mankoff, Bob</creatorcontrib><creatorcontrib>Nowak, Robert</creatorcontrib><creatorcontrib>Ervin Tanczos</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>ProQuest - Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Mankoff, Bob</au><au>Nowak, Robert</au><au>Ervin Tanczos</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>A KL-LUCB Bandit Algorithm for Large-Scale Crowdsourcing</atitle><jtitle>arXiv.org</jtitle><date>2017-09-11</date><risdate>2017</risdate><eissn>2331-8422</eissn><abstract>This paper focuses on best-arm identification in multi-armed bandits with bounded rewards. We develop an algorithm that is a fusion of lil-UCB and KL-LUCB, offering the best qualities of the two algorithms in one method. This is achieved by proving a novel anytime confidence bound for the mean of bounded distributions, which is the analogue of the LIL-type bounds recently developed for sub-Gaussian distributions. We corroborate our theoretical results with numerical experiments based on the New Yorker Cartoon Caption Contest.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2017-09
issn 2331-8422
language eng
recordid cdi_proquest_journals_2076865905
source ProQuest - Publicly Available Content Database
subjects Algorithms
Crowdsourcing
title A KL-LUCB Bandit Algorithm for Large-Scale Crowdsourcing
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-27T08%3A50%3A30IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=A%20KL-LUCB%20Bandit%20Algorithm%20for%20Large-Scale%20Crowdsourcing&rft.jtitle=arXiv.org&rft.au=Mankoff,%20Bob&rft.date=2017-09-11&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2076865905%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-proquest_journals_20768659053%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2076865905&rft_id=info:pmid/&rfr_iscdi=true