Automatic Data Augmentation by Upper Confidence Bounds for Deep Reinforcement Learning

In visual reinforcement learning (RL), various approaches succeeded to improve data efficiency. However, the approaches fail to show generalization capabilities if different colors or backgrounds are applied to its environment. The lack of generalization capabilities can hinder the use of RL in real...

Full description

Saved in:
Bibliographic Details
Main Authors: Gil, Yoonhee, Baek, Jongchan, Park, Jonghyuk, Han, Soohee
Format: Conference Proceeding
Language:English
Subjects:
Automatic data augmentation
Automation
Colored noise
Continuous control
Control systems
Image color analysis
Reinforcement learning
Upper confidence bounds
Visualization
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 1203
container_issue
container_start_page 1199
container_title
container_volume
creator Gil, Yoonhee
Baek, Jongchan
Park, Jonghyuk
Han, Soohee
description In visual reinforcement learning (RL), various approaches succeeded to improve data efficiency. However, the approaches fail to show generalization capabilities if different colors or backgrounds are applied to its environment. The lack of generalization capabilities can hinder the use of RL in real-world environment, which contains lot of distractions and noises. In this paper, a novel automatic data augmentation method that can improve generalization capabilities of an RL agent. In the experiments, the proposed method shows better generalization capabilities than other approaches. These results provide a simple automatic data augmentation method for RL that can improve generalization capabilities.
doi_str_mv 10.23919/ICCAS52745.2021.9649771
format conference_proceeding
fullrecord <record><control><sourceid>ieee_CHZPO</sourceid><recordid>TN_cdi_ieee_primary_9649771</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>9649771</ieee_id><sourcerecordid>9649771</sourcerecordid><originalsourceid>FETCH-LOGICAL-i203t-88b54a2fe5a6e5858b307f191e456237c3bef1c01b9ba9231dbcefe37df13f773</originalsourceid><addsrcrecordid>eNotkM1OwzAQhA0SEgX6BFz8Aine3TiOjyHlp1IlJKBcKztZV0bUiZL00LeniJ5G80nzHUYICWqBZME-rOq6-tBocr1AhbCwRW6NgQtxU1pLCBrBXooZFjlmZBVci_k4fiulCFWuinImvqrD1O3dFBu5dJOT1WG35zSdQJekP8pN3_Mg6y6F2HJqWD52h9SOMnSDXDL38p1jOpWG_2ZyzW5IMe3uxFVwPyPPz3krNs9Pn_Vrtn57WdXVOouoaMrK0uvcYWDtCtalLj0pE8AC57pAMg15DtAo8NY7iwStbzgwmTYABWPoVtz_eyMzb_sh7t1w3J5voF88C1NE</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Automatic Data Augmentation by Upper Confidence Bounds for Deep Reinforcement Learning</title><source>IEEE Xplore All Conference Series</source><creator>Gil, Yoonhee ; Baek, Jongchan ; Park, Jonghyuk ; Han, Soohee</creator><creatorcontrib>Gil, Yoonhee ; Baek, Jongchan ; Park, Jonghyuk ; Han, Soohee</creatorcontrib><description>In visual reinforcement learning (RL), various approaches succeeded to improve data efficiency. However, the approaches fail to show generalization capabilities if different colors or backgrounds are applied to its environment. The lack of generalization capabilities can hinder the use of RL in real-world environment, which contains lot of distractions and noises. In this paper, a novel automatic data augmentation method that can improve generalization capabilities of an RL agent. In the experiments, the proposed method shows better generalization capabilities than other approaches. These results provide a simple automatic data augmentation method for RL that can improve generalization capabilities.</description><identifier>EISSN: 2642-3901</identifier><identifier>EISBN: 8993215219</identifier><identifier>EISBN: 9788993215212</identifier><identifier>DOI: 10.23919/ICCAS52745.2021.9649771</identifier><language>eng</language><publisher>ICROS</publisher><subject>Automatic data augmentation ; Automation ; Colored noise ; Continuous control ; Control systems ; Image color analysis ; Reinforcement learning ; Upper confidence bounds ; Visualization</subject><ispartof>2021 21st International Conference on Control, Automation and Systems (ICCAS), 2021, p.1199-1203</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/9649771$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,23928,23929,25138,27923,54553,54930</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/9649771$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Gil, Yoonhee</creatorcontrib><creatorcontrib>Baek, Jongchan</creatorcontrib><creatorcontrib>Park, Jonghyuk</creatorcontrib><creatorcontrib>Han, Soohee</creatorcontrib><title>Automatic Data Augmentation by Upper Confidence Bounds for Deep Reinforcement Learning</title><title>2021 21st International Conference on Control, Automation and Systems (ICCAS)</title><addtitle>ICCAS</addtitle><description>In visual reinforcement learning (RL), various approaches succeeded to improve data efficiency. However, the approaches fail to show generalization capabilities if different colors or backgrounds are applied to its environment. The lack of generalization capabilities can hinder the use of RL in real-world environment, which contains lot of distractions and noises. In this paper, a novel automatic data augmentation method that can improve generalization capabilities of an RL agent. In the experiments, the proposed method shows better generalization capabilities than other approaches. These results provide a simple automatic data augmentation method for RL that can improve generalization capabilities.</description><subject>Automatic data augmentation</subject><subject>Automation</subject><subject>Colored noise</subject><subject>Continuous control</subject><subject>Control systems</subject><subject>Image color analysis</subject><subject>Reinforcement learning</subject><subject>Upper confidence bounds</subject><subject>Visualization</subject><issn>2642-3901</issn><isbn>8993215219</isbn><isbn>9788993215212</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2021</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotkM1OwzAQhA0SEgX6BFz8Aine3TiOjyHlp1IlJKBcKztZV0bUiZL00LeniJ5G80nzHUYICWqBZME-rOq6-tBocr1AhbCwRW6NgQtxU1pLCBrBXooZFjlmZBVci_k4fiulCFWuinImvqrD1O3dFBu5dJOT1WG35zSdQJekP8pN3_Mg6y6F2HJqWD52h9SOMnSDXDL38p1jOpWG_2ZyzW5IMe3uxFVwPyPPz3krNs9Pn_Vrtn57WdXVOouoaMrK0uvcYWDtCtalLj0pE8AC57pAMg15DtAo8NY7iwStbzgwmTYABWPoVtz_eyMzb_sh7t1w3J5voF88C1NE</recordid><startdate>20211012</startdate><enddate>20211012</enddate><creator>Gil, Yoonhee</creator><creator>Baek, Jongchan</creator><creator>Park, Jonghyuk</creator><creator>Han, Soohee</creator><general>ICROS</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>20211012</creationdate><title>Automatic Data Augmentation by Upper Confidence Bounds for Deep Reinforcement Learning</title><author>Gil, Yoonhee ; Baek, Jongchan ; Park, Jonghyuk ; Han, Soohee</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i203t-88b54a2fe5a6e5858b307f191e456237c3bef1c01b9ba9231dbcefe37df13f773</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Automatic data augmentation</topic><topic>Automation</topic><topic>Colored noise</topic><topic>Continuous control</topic><topic>Control systems</topic><topic>Image color analysis</topic><topic>Reinforcement learning</topic><topic>Upper confidence bounds</topic><topic>Visualization</topic><toplevel>online_resources</toplevel><creatorcontrib>Gil, Yoonhee</creatorcontrib><creatorcontrib>Baek, Jongchan</creatorcontrib><creatorcontrib>Park, Jonghyuk</creatorcontrib><creatorcontrib>Han, Soohee</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Xplore</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Gil, Yoonhee</au><au>Baek, Jongchan</au><au>Park, Jonghyuk</au><au>Han, Soohee</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Automatic Data Augmentation by Upper Confidence Bounds for Deep Reinforcement Learning</atitle><btitle>2021 21st International Conference on Control, Automation and Systems (ICCAS)</btitle><stitle>ICCAS</stitle><date>2021-10-12</date><risdate>2021</risdate><spage>1199</spage><epage>1203</epage><pages>1199-1203</pages><eissn>2642-3901</eissn><eisbn>8993215219</eisbn><eisbn>9788993215212</eisbn><abstract>In visual reinforcement learning (RL), various approaches succeeded to improve data efficiency. However, the approaches fail to show generalization capabilities if different colors or backgrounds are applied to its environment. The lack of generalization capabilities can hinder the use of RL in real-world environment, which contains lot of distractions and noises. In this paper, a novel automatic data augmentation method that can improve generalization capabilities of an RL agent. In the experiments, the proposed method shows better generalization capabilities than other approaches. These results provide a simple automatic data augmentation method for RL that can improve generalization capabilities.</abstract><pub>ICROS</pub><doi>10.23919/ICCAS52745.2021.9649771</doi><tpages>5</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier EISSN: 2642-3901
ispartof 2021 21st International Conference on Control, Automation and Systems (ICCAS), 2021, p.1199-1203
issn 2642-3901
language eng
recordid cdi_ieee_primary_9649771
source IEEE Xplore All Conference Series
subjects Automatic data augmentation
Automation
Colored noise
Continuous control
Control systems
Image color analysis
Reinforcement learning
Upper confidence bounds
Visualization
title Automatic Data Augmentation by Upper Confidence Bounds for Deep Reinforcement Learning
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-13T22%3A07%3A17IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_CHZPO&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Automatic%20Data%20Augmentation%20by%20Upper%20Confidence%20Bounds%20for%20Deep%20Reinforcement%20Learning&rft.btitle=2021%2021st%20International%20Conference%20on%20Control,%20Automation%20and%20Systems%20(ICCAS)&rft.au=Gil,%20Yoonhee&rft.date=2021-10-12&rft.spage=1199&rft.epage=1203&rft.pages=1199-1203&rft.eissn=2642-3901&rft_id=info:doi/10.23919/ICCAS52745.2021.9649771&rft.eisbn=8993215219&rft.eisbn_list=9788993215212&rft_dat=%3Cieee_CHZPO%3E9649771%3C/ieee_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i203t-88b54a2fe5a6e5858b307f191e456237c3bef1c01b9ba9231dbcefe37df13f773%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=9649771&rfr_iscdi=true