Continual Robot Learning with Constructive Neural Networks
In this paper, we present an approach for combining reinforcement learning, learning by imitation, and incremental hierarchical development. We apply this approach to a realistic simulated mobile robot that learns to perform a navigation task by imitating the movements of a teacher and then continue...
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Buchkapitel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 108 |
|---|---|
| container_issue | |
| container_start_page | 95 |
| container_title | |
| container_volume | 1545 |
| creator | Großmann, Axel Poli, Riccardo |
| description | In this paper, we present an approach for combining reinforcement learning, learning by imitation, and incremental hierarchical development. We apply this approach to a realistic simulated mobile robot that learns to perform a navigation task by imitating the movements of a teacher and then continues to learn by receiving reinforcement. The behaviours of the robot are represented as sensation-action rules in a constructive high-order neural network. Preliminary experiments are reported which show that incremental, hierarchical development, bootstrapped by imitative learning, allows the robot to adapt to changes in its environment during its entire lifetime very efficiently, even if only delayed reinforcements are given. |
| doi_str_mv | 10.1007/3-540-49240-2_7 |
| format | Book Chapter |
| fullrecord | <record><control><sourceid>proquest_pasca</sourceid><recordid>TN_cdi_pascalfrancis_primary_1574080</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>EBC3072076_12_103</sourcerecordid><originalsourceid>FETCH-LOGICAL-p266t-e37623f6ae3e17d64de14e23c9cdfa8c7f2956c1b1b543df4db7f87b25ab1c5e3</originalsourceid><addsrcrecordid>eNotUMtOwzAQNE8RSs9cc-BqsL1-JNxQxUuqioTgbDmO04aWpNgOFX-P-9jDrnZndqQZhK4puaWEqDvAghPMS5Y60-oIXUI67HZ2jDIqKcUAvDxB41IVW0wKXhA4RRkBwnCpOJyjrBSFYJSWxQUah_BFUgHjwIsM3U_6LrbdYFb5e1_1MZ8647u2m-ebNi7yhIboBxvbX5fP3OATb-bipvfLcIXOGrMKbnyYI_T59PgxecHTt-fXycMUr5mUETtQkkEjjQNHVS157Sh3DGxp68YUVjWsFNLSilaCQ93wulJNoSomTEWtcDBCN3vdtQnWrBpvOtsGvfbtt_F_mgrFSfI8QnhPCwnp5s7rqu-XQVOit1lq0CkevctOpywTHw6yvv8ZXIjabR-s62JyaRdmHZ0PGohiRElNWZIB-AebPXG5</addsrcrecordid><sourcetype>Index Database</sourcetype><iscdi>true</iscdi><recordtype>book_chapter</recordtype><pqid>EBC3072076_12_103</pqid></control><display><type>book_chapter</type><title>Continual Robot Learning with Constructive Neural Networks</title><source>Springer Books</source><creator>Großmann, Axel ; Poli, Riccardo</creator><contributor>Demiris, John ; van Leeuwen, J ; Birk, Andreas ; Demiris, John ; Birk, Andreas</contributor><creatorcontrib>Großmann, Axel ; Poli, Riccardo ; Demiris, John ; van Leeuwen, J ; Birk, Andreas ; Demiris, John ; Birk, Andreas</creatorcontrib><description>In this paper, we present an approach for combining reinforcement learning, learning by imitation, and incremental hierarchical development. We apply this approach to a realistic simulated mobile robot that learns to perform a navigation task by imitating the movements of a teacher and then continues to learn by receiving reinforcement. The behaviours of the robot are represented as sensation-action rules in a constructive high-order neural network. Preliminary experiments are reported which show that incremental, hierarchical development, bootstrapped by imitative learning, allows the robot to adapt to changes in its environment during its entire lifetime very efficiently, even if only delayed reinforcements are given.</description><identifier>ISSN: 0302-9743</identifier><identifier>ISBN: 9783540654803</identifier><identifier>ISBN: 3540654801</identifier><identifier>EISSN: 1611-3349</identifier><identifier>EISBN: 3540492402</identifier><identifier>EISBN: 9783540492405</identifier><identifier>DOI: 10.1007/3-540-49240-2_7</identifier><identifier>OCLC: 958521198</identifier><identifier>LCCallNum: TA1-2040</identifier><language>eng</language><publisher>Germany: Springer Berlin / Heidelberg</publisher><subject>Applied sciences ; Artificial intelligence ; Computer science; control theory; systems ; Control theory. Systems ; Exact sciences and technology ; Goal Position ; Learning and adaptive systems ; Learning Task ; Mobile Robot ; Navigation Task ; Robotics ; Supervise Learning Algorithm</subject><ispartof>Learning Robots, 1998, Vol.1545, p.95-108</ispartof><rights>Springer-Verlag Berlin Heidelberg 1998</rights><rights>1999 INIST-CNRS</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><relation>Lecture Notes in Computer Science</relation></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Uhttps://ebookcentral.proquest.com/covers/3072076-l.jpg</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/3-540-49240-2_7$$EPDF$$P50$$Gspringer$$H</linktopdf><linktohtml>$$Uhttps://link.springer.com/10.1007/3-540-49240-2_7$$EHTML$$P50$$Gspringer$$H</linktohtml><link.rule.ids>309,310,775,776,780,785,786,789,27902,38232,41418,42487</link.rule.ids><backlink>$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=1574080$$DView record in Pascal Francis$$Hfree_for_read</backlink></links><search><contributor>Demiris, John</contributor><contributor>van Leeuwen, J</contributor><contributor>Birk, Andreas</contributor><contributor>Demiris, John</contributor><contributor>Birk, Andreas</contributor><creatorcontrib>Großmann, Axel</creatorcontrib><creatorcontrib>Poli, Riccardo</creatorcontrib><title>Continual Robot Learning with Constructive Neural Networks</title><title>Learning Robots</title><description>In this paper, we present an approach for combining reinforcement learning, learning by imitation, and incremental hierarchical development. We apply this approach to a realistic simulated mobile robot that learns to perform a navigation task by imitating the movements of a teacher and then continues to learn by receiving reinforcement. The behaviours of the robot are represented as sensation-action rules in a constructive high-order neural network. Preliminary experiments are reported which show that incremental, hierarchical development, bootstrapped by imitative learning, allows the robot to adapt to changes in its environment during its entire lifetime very efficiently, even if only delayed reinforcements are given.</description><subject>Applied sciences</subject><subject>Artificial intelligence</subject><subject>Computer science; control theory; systems</subject><subject>Control theory. Systems</subject><subject>Exact sciences and technology</subject><subject>Goal Position</subject><subject>Learning and adaptive systems</subject><subject>Learning Task</subject><subject>Mobile Robot</subject><subject>Navigation Task</subject><subject>Robotics</subject><subject>Supervise Learning Algorithm</subject><issn>0302-9743</issn><issn>1611-3349</issn><isbn>9783540654803</isbn><isbn>3540654801</isbn><isbn>3540492402</isbn><isbn>9783540492405</isbn><fulltext>true</fulltext><rsrctype>book_chapter</rsrctype><creationdate>1998</creationdate><recordtype>book_chapter</recordtype><recordid>eNotUMtOwzAQNE8RSs9cc-BqsL1-JNxQxUuqioTgbDmO04aWpNgOFX-P-9jDrnZndqQZhK4puaWEqDvAghPMS5Y60-oIXUI67HZ2jDIqKcUAvDxB41IVW0wKXhA4RRkBwnCpOJyjrBSFYJSWxQUah_BFUgHjwIsM3U_6LrbdYFb5e1_1MZ8647u2m-ebNi7yhIboBxvbX5fP3OATb-bipvfLcIXOGrMKbnyYI_T59PgxecHTt-fXycMUr5mUETtQkkEjjQNHVS157Sh3DGxp68YUVjWsFNLSilaCQ93wulJNoSomTEWtcDBCN3vdtQnWrBpvOtsGvfbtt_F_mgrFSfI8QnhPCwnp5s7rqu-XQVOit1lq0CkevctOpywTHw6yvv8ZXIjabR-s62JyaRdmHZ0PGohiRElNWZIB-AebPXG5</recordid><startdate>19980101</startdate><enddate>19980101</enddate><creator>Großmann, Axel</creator><creator>Poli, Riccardo</creator><general>Springer Berlin / Heidelberg</general><general>Springer Berlin Heidelberg</general><general>Springer</general><scope>FFUUA</scope><scope>IQODW</scope></search><sort><creationdate>19980101</creationdate><title>Continual Robot Learning with Constructive Neural Networks</title><author>Großmann, Axel ; Poli, Riccardo</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-p266t-e37623f6ae3e17d64de14e23c9cdfa8c7f2956c1b1b543df4db7f87b25ab1c5e3</frbrgroupid><rsrctype>book_chapters</rsrctype><prefilter>book_chapters</prefilter><language>eng</language><creationdate>1998</creationdate><topic>Applied sciences</topic><topic>Artificial intelligence</topic><topic>Computer science; control theory; systems</topic><topic>Control theory. Systems</topic><topic>Exact sciences and technology</topic><topic>Goal Position</topic><topic>Learning and adaptive systems</topic><topic>Learning Task</topic><topic>Mobile Robot</topic><topic>Navigation Task</topic><topic>Robotics</topic><topic>Supervise Learning Algorithm</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Großmann, Axel</creatorcontrib><creatorcontrib>Poli, Riccardo</creatorcontrib><collection>ProQuest Ebook Central - Book Chapters - Demo use only</collection><collection>Pascal-Francis</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Großmann, Axel</au><au>Poli, Riccardo</au><au>Demiris, John</au><au>van Leeuwen, J</au><au>Birk, Andreas</au><au>Demiris, John</au><au>Birk, Andreas</au><format>book</format><genre>bookitem</genre><ristype>CHAP</ristype><atitle>Continual Robot Learning with Constructive Neural Networks</atitle><btitle>Learning Robots</btitle><seriestitle>Lecture Notes in Computer Science</seriestitle><date>1998-01-01</date><risdate>1998</risdate><volume>1545</volume><spage>95</spage><epage>108</epage><pages>95-108</pages><issn>0302-9743</issn><eissn>1611-3349</eissn><isbn>9783540654803</isbn><isbn>3540654801</isbn><eisbn>3540492402</eisbn><eisbn>9783540492405</eisbn><abstract>In this paper, we present an approach for combining reinforcement learning, learning by imitation, and incremental hierarchical development. We apply this approach to a realistic simulated mobile robot that learns to perform a navigation task by imitating the movements of a teacher and then continues to learn by receiving reinforcement. The behaviours of the robot are represented as sensation-action rules in a constructive high-order neural network. Preliminary experiments are reported which show that incremental, hierarchical development, bootstrapped by imitative learning, allows the robot to adapt to changes in its environment during its entire lifetime very efficiently, even if only delayed reinforcements are given.</abstract><cop>Germany</cop><pub>Springer Berlin / Heidelberg</pub><doi>10.1007/3-540-49240-2_7</doi><oclcid>958521198</oclcid><tpages>14</tpages></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0302-9743 |
| ispartof | Learning Robots, 1998, Vol.1545, p.95-108 |
| issn | 0302-9743 1611-3349 |
| language | eng |
| recordid | cdi_pascalfrancis_primary_1574080 |
| source | Springer Books |
| subjects | Applied sciences Artificial intelligence Computer science control theory systems Control theory. Systems Exact sciences and technology Goal Position Learning and adaptive systems Learning Task Mobile Robot Navigation Task Robotics Supervise Learning Algorithm |
| title | Continual Robot Learning with Constructive Neural Networks |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-04T22%3A15%3A07IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pasca&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=bookitem&rft.atitle=Continual%20Robot%20Learning%20with%20Constructive%20Neural%20Networks&rft.btitle=Learning%20Robots&rft.au=Gro%C3%9Fmann,%20Axel&rft.date=1998-01-01&rft.volume=1545&rft.spage=95&rft.epage=108&rft.pages=95-108&rft.issn=0302-9743&rft.eissn=1611-3349&rft.isbn=9783540654803&rft.isbn_list=3540654801&rft_id=info:doi/10.1007/3-540-49240-2_7&rft_dat=%3Cproquest_pasca%3EEBC3072076_12_103%3C/proquest_pasca%3E%3Curl%3E%3C/url%3E&rft.eisbn=3540492402&rft.eisbn_list=9783540492405&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=EBC3072076_12_103&rft_id=info:pmid/&rfr_iscdi=true |