CINet: A Learning Based Approach to Incremental Context Modeling in Robots

There have been several attempts at modeling context in robots. However, either these attempts assume a fixed number of contexts or use a rule-based approach to determine when to increment the number of contexts. In this paper, we pose the task of when to increment as a learning problem, which we so...

Full description

Saved in:
Bibliographic Details
Main Authors: Irmak Dogan, Fethiye, Bozcan, Ilker, Celik, Mehmet, Kalkan, Sinan
Format: Conference Proceeding
Language:English
Subjects:
Computational modeling
Context modeling
Recurrent neural networks
Resource management
Robots
Testing
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 4646
container_issue
container_start_page 4641
container_title
container_volume
creator Irmak Dogan, Fethiye
Bozcan, Ilker
Celik, Mehmet
Kalkan, Sinan
description There have been several attempts at modeling context in robots. However, either these attempts assume a fixed number of contexts or use a rule-based approach to determine when to increment the number of contexts. In this paper, we pose the task of when to increment as a learning problem, which we solve using a Recurrent Neural Network. We show that the network successfully (with 98% testing accuracy) learns to predict when to increment, and demonstrate, in a scene modeling problem (where the correct number of contexts is not known), that the robot increments the number of contexts in an expected manner (i.e., the entropy of the system is reduced). We also present how the incremental model can be used for various scene reasoning tasks.
doi_str_mv 10.1109/IROS.2018.8593633
format conference_proceeding
fullrecord <record><control><sourceid>ieee_CHZPO</sourceid><recordid>TN_cdi_ieee_primary_8593633</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>8593633</ieee_id><sourcerecordid>8593633</sourcerecordid><originalsourceid>FETCH-LOGICAL-i241t-5c41c8c11623cfc4e4d7ca463558e9f465967247db53a959bbb69848ab837ef83</originalsourceid><addsrcrecordid>eNotj81KxDAURqMgOIx9AHGTF2jNf27c1aJjpTow6npI01utdNrSZqFvr-KsPg4cDnyEXHKWcc7cdbnbvmSCcchAO2mkPCGJs8C1BAPMKXZKVuKXUgbGnJNkWT4ZY8KABKtW5LEonzHe0JxW6OehG97prV-wofk0zaMPHzSOtBzCjAccou9pMQ4RvyJ9Ghvs__RuoLuxHuNyQc5a3y-YHHdN3u7vXouHtNpuyiKv0k4oHlMdFA8QODdChjYoVI0NXhmpNaBrldHOWKFsU2vpnXZ1XRsHCnwN0mILck2u_rsdIu6nuTv4-Xt_fC9_ABUATG4</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>CINet: A Learning Based Approach to Incremental Context Modeling in Robots</title><source>IEEE Xplore All Conference Series</source><creator>Irmak Dogan, Fethiye ; Bozcan, Ilker ; Celik, Mehmet ; Kalkan, Sinan</creator><creatorcontrib>Irmak Dogan, Fethiye ; Bozcan, Ilker ; Celik, Mehmet ; Kalkan, Sinan</creatorcontrib><description>There have been several attempts at modeling context in robots. However, either these attempts assume a fixed number of contexts or use a rule-based approach to determine when to increment the number of contexts. In this paper, we pose the task of when to increment as a learning problem, which we solve using a Recurrent Neural Network. We show that the network successfully (with 98% testing accuracy) learns to predict when to increment, and demonstrate, in a scene modeling problem (where the correct number of contexts is not known), that the robot increments the number of contexts in an expected manner (i.e., the entropy of the system is reduced). We also present how the incremental model can be used for various scene reasoning tasks.</description><identifier>EISSN: 2153-0866</identifier><identifier>EISBN: 9781538680940</identifier><identifier>EISBN: 1538680947</identifier><identifier>DOI: 10.1109/IROS.2018.8593633</identifier><language>eng</language><publisher>IEEE</publisher><subject>Computational modeling ; Context modeling ; Recurrent neural networks ; Resource management ; Robots ; Testing ; Training</subject><ispartof>2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018, p.4641-4646</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/8593633$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,27925,54555,54932</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/8593633$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Irmak Dogan, Fethiye</creatorcontrib><creatorcontrib>Bozcan, Ilker</creatorcontrib><creatorcontrib>Celik, Mehmet</creatorcontrib><creatorcontrib>Kalkan, Sinan</creatorcontrib><title>CINet: A Learning Based Approach to Incremental Context Modeling in Robots</title><title>2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)</title><addtitle>IROS</addtitle><description>There have been several attempts at modeling context in robots. However, either these attempts assume a fixed number of contexts or use a rule-based approach to determine when to increment the number of contexts. In this paper, we pose the task of when to increment as a learning problem, which we solve using a Recurrent Neural Network. We show that the network successfully (with 98% testing accuracy) learns to predict when to increment, and demonstrate, in a scene modeling problem (where the correct number of contexts is not known), that the robot increments the number of contexts in an expected manner (i.e., the entropy of the system is reduced). We also present how the incremental model can be used for various scene reasoning tasks.</description><subject>Computational modeling</subject><subject>Context modeling</subject><subject>Recurrent neural networks</subject><subject>Resource management</subject><subject>Robots</subject><subject>Testing</subject><subject>Training</subject><issn>2153-0866</issn><isbn>9781538680940</isbn><isbn>1538680947</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2018</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotj81KxDAURqMgOIx9AHGTF2jNf27c1aJjpTow6npI01utdNrSZqFvr-KsPg4cDnyEXHKWcc7cdbnbvmSCcchAO2mkPCGJs8C1BAPMKXZKVuKXUgbGnJNkWT4ZY8KABKtW5LEonzHe0JxW6OehG97prV-wofk0zaMPHzSOtBzCjAccou9pMQ4RvyJ9Ghvs__RuoLuxHuNyQc5a3y-YHHdN3u7vXouHtNpuyiKv0k4oHlMdFA8QODdChjYoVI0NXhmpNaBrldHOWKFsU2vpnXZ1XRsHCnwN0mILck2u_rsdIu6nuTv4-Xt_fC9_ABUATG4</recordid><startdate>201810</startdate><enddate>201810</enddate><creator>Irmak Dogan, Fethiye</creator><creator>Bozcan, Ilker</creator><creator>Celik, Mehmet</creator><creator>Kalkan, Sinan</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope></search><sort><creationdate>201810</creationdate><title>CINet: A Learning Based Approach to Incremental Context Modeling in Robots</title><author>Irmak Dogan, Fethiye ; Bozcan, Ilker ; Celik, Mehmet ; Kalkan, Sinan</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i241t-5c41c8c11623cfc4e4d7ca463558e9f465967247db53a959bbb69848ab837ef83</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Computational modeling</topic><topic>Context modeling</topic><topic>Recurrent neural networks</topic><topic>Resource management</topic><topic>Robots</topic><topic>Testing</topic><topic>Training</topic><toplevel>online_resources</toplevel><creatorcontrib>Irmak Dogan, Fethiye</creatorcontrib><creatorcontrib>Bozcan, Ilker</creatorcontrib><creatorcontrib>Celik, Mehmet</creatorcontrib><creatorcontrib>Kalkan, Sinan</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan (POP) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Xplore (Online service)</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Irmak Dogan, Fethiye</au><au>Bozcan, Ilker</au><au>Celik, Mehmet</au><au>Kalkan, Sinan</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>CINet: A Learning Based Approach to Incremental Context Modeling in Robots</atitle><btitle>2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)</btitle><stitle>IROS</stitle><date>2018-10</date><risdate>2018</risdate><spage>4641</spage><epage>4646</epage><pages>4641-4646</pages><eissn>2153-0866</eissn><eisbn>9781538680940</eisbn><eisbn>1538680947</eisbn><abstract>There have been several attempts at modeling context in robots. However, either these attempts assume a fixed number of contexts or use a rule-based approach to determine when to increment the number of contexts. In this paper, we pose the task of when to increment as a learning problem, which we solve using a Recurrent Neural Network. We show that the network successfully (with 98% testing accuracy) learns to predict when to increment, and demonstrate, in a scene modeling problem (where the correct number of contexts is not known), that the robot increments the number of contexts in an expected manner (i.e., the entropy of the system is reduced). We also present how the incremental model can be used for various scene reasoning tasks.</abstract><pub>IEEE</pub><doi>10.1109/IROS.2018.8593633</doi><tpages>6</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier EISSN: 2153-0866
ispartof 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018, p.4641-4646
issn 2153-0866
language eng
recordid cdi_ieee_primary_8593633
source IEEE Xplore All Conference Series
subjects Computational modeling
Context modeling
Recurrent neural networks
Resource management
Robots
Testing
Training
title CINet: A Learning Based Approach to Incremental Context Modeling in Robots
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-27T17%3A18%3A15IST&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=CINet:%20A%20Learning%20Based%20Approach%20to%20Incremental%20Context%20Modeling%20in%20Robots&rft.btitle=2018%20IEEE/RSJ%20International%20Conference%20on%20Intelligent%20Robots%20and%20Systems%20(IROS)&rft.au=Irmak%20Dogan,%20Fethiye&rft.date=2018-10&rft.spage=4641&rft.epage=4646&rft.pages=4641-4646&rft.eissn=2153-0866&rft_id=info:doi/10.1109/IROS.2018.8593633&rft.eisbn=9781538680940&rft.eisbn_list=1538680947&rft_dat=%3Cieee_CHZPO%3E8593633%3C/ieee_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i241t-5c41c8c11623cfc4e4d7ca463558e9f465967247db53a959bbb69848ab837ef83%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=8593633&rfr_iscdi=true