Mastering the Dungeon: Grounded Language Learning by Mechanical Turker Descent
Contrary to most natural language processing research, which makes use of static datasets, humans learn language interactively, grounded in an environment. In this work we propose an interactive learning procedure called Mechanical Turker Descent (MTD) and use it to train agents to execute natural l...
Gespeichert in:
| Veröffentlicht in: | arXiv.org 2018-04 |
|---|---|
| Hauptverfasser: | , , , , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | |
| container_title | arXiv.org |
| container_volume | |
| creator | Yang, Zhilin Zhang, Saizheng Urbanek, Jack Feng, Will Miller, Alexander H Szlam, Arthur Kiela, Douwe Weston, Jason |
| description | Contrary to most natural language processing research, which makes use of static datasets, humans learn language interactively, grounded in an environment. In this work we propose an interactive learning procedure called Mechanical Turker Descent (MTD) and use it to train agents to execute natural language commands grounded in a fantasy text adventure game. In MTD, Turkers compete to train better agents in the short term, and collaborate by sharing their agents' skills in the long term. This results in a gamified, engaging experience for the Turkers and a better quality teaching signal for the agents compared to static datasets, as the Turkers naturally adapt the training data to the agent's abilities. |
| format | Article |
| fullrecord | <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2072007286</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2072007286</sourcerecordid><originalsourceid>FETCH-proquest_journals_20720072863</originalsourceid><addsrcrecordid>eNqNissKgkAUQIcgSMp_uNBamMZ80DZ7LLRVe5n0Nmpyp-ax6O8z6ANaHM7inBkLRBxvonwrxIKF1g6cc5FmIknigF0qaR2anhS4DqHwpFDTDk5Ge2qxhVKS8lIhlCgNfb_bGypsOkl9I0e4evNAAwXaBsmt2PwuR4vhz0u2Ph6u-3P0NPrl0bp60N7QlGrBM8En8jT-7_oAnSs9uw</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2072007286</pqid></control><display><type>article</type><title>Mastering the Dungeon: Grounded Language Learning by Mechanical Turker Descent</title><source>Free E- Journals</source><creator>Yang, Zhilin ; Zhang, Saizheng ; Urbanek, Jack ; Feng, Will ; Miller, Alexander H ; Szlam, Arthur ; Kiela, Douwe ; Weston, Jason</creator><creatorcontrib>Yang, Zhilin ; Zhang, Saizheng ; Urbanek, Jack ; Feng, Will ; Miller, Alexander H ; Szlam, Arthur ; Kiela, Douwe ; Weston, Jason</creatorcontrib><description>Contrary to most natural language processing research, which makes use of static datasets, humans learn language interactively, grounded in an environment. In this work we propose an interactive learning procedure called Mechanical Turker Descent (MTD) and use it to train agents to execute natural language commands grounded in a fantasy text adventure game. In MTD, Turkers compete to train better agents in the short term, and collaborate by sharing their agents' skills in the long term. This results in a gamified, engaging experience for the Turkers and a better quality teaching signal for the agents compared to static datasets, as the Turkers naturally adapt the training data to the agent's abilities.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Datasets ; Descent ; Interactive learning ; Mastering ; Natural language ; Natural language processing ; Signal quality</subject><ispartof>arXiv.org, 2018-04</ispartof><rights>2018. 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><link.rule.ids>781,785</link.rule.ids></links><search><creatorcontrib>Yang, Zhilin</creatorcontrib><creatorcontrib>Zhang, Saizheng</creatorcontrib><creatorcontrib>Urbanek, Jack</creatorcontrib><creatorcontrib>Feng, Will</creatorcontrib><creatorcontrib>Miller, Alexander H</creatorcontrib><creatorcontrib>Szlam, Arthur</creatorcontrib><creatorcontrib>Kiela, Douwe</creatorcontrib><creatorcontrib>Weston, Jason</creatorcontrib><title>Mastering the Dungeon: Grounded Language Learning by Mechanical Turker Descent</title><title>arXiv.org</title><description>Contrary to most natural language processing research, which makes use of static datasets, humans learn language interactively, grounded in an environment. In this work we propose an interactive learning procedure called Mechanical Turker Descent (MTD) and use it to train agents to execute natural language commands grounded in a fantasy text adventure game. In MTD, Turkers compete to train better agents in the short term, and collaborate by sharing their agents' skills in the long term. This results in a gamified, engaging experience for the Turkers and a better quality teaching signal for the agents compared to static datasets, as the Turkers naturally adapt the training data to the agent's abilities.</description><subject>Datasets</subject><subject>Descent</subject><subject>Interactive learning</subject><subject>Mastering</subject><subject>Natural language</subject><subject>Natural language processing</subject><subject>Signal quality</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2018</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><recordid>eNqNissKgkAUQIcgSMp_uNBamMZ80DZ7LLRVe5n0Nmpyp-ax6O8z6ANaHM7inBkLRBxvonwrxIKF1g6cc5FmIknigF0qaR2anhS4DqHwpFDTDk5Ge2qxhVKS8lIhlCgNfb_bGypsOkl9I0e4evNAAwXaBsmt2PwuR4vhz0u2Ph6u-3P0NPrl0bp60N7QlGrBM8En8jT-7_oAnSs9uw</recordid><startdate>20180416</startdate><enddate>20180416</enddate><creator>Yang, Zhilin</creator><creator>Zhang, Saizheng</creator><creator>Urbanek, Jack</creator><creator>Feng, Will</creator><creator>Miller, Alexander H</creator><creator>Szlam, Arthur</creator><creator>Kiela, Douwe</creator><creator>Weston, Jason</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>20180416</creationdate><title>Mastering the Dungeon: Grounded Language Learning by Mechanical Turker Descent</title><author>Yang, Zhilin ; Zhang, Saizheng ; Urbanek, Jack ; Feng, Will ; Miller, Alexander H ; Szlam, Arthur ; Kiela, Douwe ; Weston, Jason</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_20720072863</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Datasets</topic><topic>Descent</topic><topic>Interactive learning</topic><topic>Mastering</topic><topic>Natural language</topic><topic>Natural language processing</topic><topic>Signal quality</topic><toplevel>online_resources</toplevel><creatorcontrib>Yang, Zhilin</creatorcontrib><creatorcontrib>Zhang, Saizheng</creatorcontrib><creatorcontrib>Urbanek, Jack</creatorcontrib><creatorcontrib>Feng, Will</creatorcontrib><creatorcontrib>Miller, Alexander H</creatorcontrib><creatorcontrib>Szlam, Arthur</creatorcontrib><creatorcontrib>Kiela, Douwe</creatorcontrib><creatorcontrib>Weston, Jason</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Access via ProQuest (Open Access)</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>Yang, Zhilin</au><au>Zhang, Saizheng</au><au>Urbanek, Jack</au><au>Feng, Will</au><au>Miller, Alexander H</au><au>Szlam, Arthur</au><au>Kiela, Douwe</au><au>Weston, Jason</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Mastering the Dungeon: Grounded Language Learning by Mechanical Turker Descent</atitle><jtitle>arXiv.org</jtitle><date>2018-04-16</date><risdate>2018</risdate><eissn>2331-8422</eissn><abstract>Contrary to most natural language processing research, which makes use of static datasets, humans learn language interactively, grounded in an environment. In this work we propose an interactive learning procedure called Mechanical Turker Descent (MTD) and use it to train agents to execute natural language commands grounded in a fantasy text adventure game. In MTD, Turkers compete to train better agents in the short term, and collaborate by sharing their agents' skills in the long term. This results in a gamified, engaging experience for the Turkers and a better quality teaching signal for the agents compared to static datasets, as the Turkers naturally adapt the training data to the agent's abilities.</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, 2018-04 |
| issn | 2331-8422 |
| language | eng |
| recordid | cdi_proquest_journals_2072007286 |
| source | Free E- Journals |
| subjects | Datasets Descent Interactive learning Mastering Natural language Natural language processing Signal quality |
| title | Mastering the Dungeon: Grounded Language Learning by Mechanical Turker Descent |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-14T18%3A40%3A40IST&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=Mastering%20the%20Dungeon:%20Grounded%20Language%20Learning%20by%20Mechanical%20Turker%20Descent&rft.jtitle=arXiv.org&rft.au=Yang,%20Zhilin&rft.date=2018-04-16&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2072007286%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2072007286&rft_id=info:pmid/&rfr_iscdi=true |