A Deep Reinforcement Learning Chatbot (Short Version)
We present MILABOT: a deep reinforcement learning chatbot developed by the Montreal Institute for Learning Algorithms (MILA) for the Amazon Alexa Prize competition. MILABOT is capable of conversing with humans on popular small talk topics through both speech and text. The system consists of an ensem...
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | |
| container_title | |
| container_volume | |
| creator | Serban, Iulian V Sankar, Chinnadhurai Germain, Mathieu Zhang, Saizheng Lin, Zhouhan Subramanian, Sandeep Kim, Taesup Pieper, Michael Chandar, Sarath Ke, Nan Rosemary Rajeswar, Sai de Brebisson, Alexandre Sotelo, Jose M. R Suhubdy, Dendi Michalski, Vincent Nguyen, Alexandre Pineau, Joelle Bengio, Yoshua |
| description | We present MILABOT: a deep reinforcement learning chatbot developed by the
Montreal Institute for Learning Algorithms (MILA) for the Amazon Alexa Prize
competition. MILABOT is capable of conversing with humans on popular small talk
topics through both speech and text. The system consists of an ensemble of
natural language generation and retrieval models, including neural network and
template-based models. By applying reinforcement learning to crowdsourced data
and real-world user interactions, the system has been trained to select an
appropriate response from the models in its ensemble. The system has been
evaluated through A/B testing with real-world users, where it performed
significantly better than other systems. The results highlight the potential of
coupling ensemble systems with deep reinforcement learning as a fruitful path
for developing real-world, open-domain conversational agents. |
| doi_str_mv | 10.48550/arxiv.1801.06700 |
| format | Article |
| fullrecord | <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1801_06700</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1801_06700</sourcerecordid><originalsourceid>FETCH-LOGICAL-a670-a983731ec10e6b5d97f99724d84d05136abdeb517db23533a69f4a9b8c4dd3283</originalsourceid><addsrcrecordid>eNotzrFOwzAUQFEvDKjwAUx4hCHBzrNje6xCgUqRKtGqa_Qcv1BL1KncqIK_B0qnu10dxu6kKJXVWjxh_oqnUlohS1EbIa6ZnvNnogN_p5iGMfe0pzTxljCnmD54s8PJjxN_WO_GPPEt5WMc0-MNuxrw80i3l87Y5mWxad6KdvW6bOZtgb_3Ap0FA5J6Kaj2OjgzOGcqFawKQkuo0QfyWprgK9AAWLtBofO2VyFAZWHG7v-3Z3d3yHGP-bv783dnP_wA_ZM-kg</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>A Deep Reinforcement Learning Chatbot (Short Version)</title><source>arXiv.org</source><creator>Serban, Iulian V ; Sankar, Chinnadhurai ; Germain, Mathieu ; Zhang, Saizheng ; Lin, Zhouhan ; Subramanian, Sandeep ; Kim, Taesup ; Pieper, Michael ; Chandar, Sarath ; Ke, Nan Rosemary ; Rajeswar, Sai ; de Brebisson, Alexandre ; Sotelo, Jose M. R ; Suhubdy, Dendi ; Michalski, Vincent ; Nguyen, Alexandre ; Pineau, Joelle ; Bengio, Yoshua</creator><creatorcontrib>Serban, Iulian V ; Sankar, Chinnadhurai ; Germain, Mathieu ; Zhang, Saizheng ; Lin, Zhouhan ; Subramanian, Sandeep ; Kim, Taesup ; Pieper, Michael ; Chandar, Sarath ; Ke, Nan Rosemary ; Rajeswar, Sai ; de Brebisson, Alexandre ; Sotelo, Jose M. R ; Suhubdy, Dendi ; Michalski, Vincent ; Nguyen, Alexandre ; Pineau, Joelle ; Bengio, Yoshua</creatorcontrib><description>We present MILABOT: a deep reinforcement learning chatbot developed by the
Montreal Institute for Learning Algorithms (MILA) for the Amazon Alexa Prize
competition. MILABOT is capable of conversing with humans on popular small talk
topics through both speech and text. The system consists of an ensemble of
natural language generation and retrieval models, including neural network and
template-based models. By applying reinforcement learning to crowdsourced data
and real-world user interactions, the system has been trained to select an
appropriate response from the models in its ensemble. The system has been
evaluated through A/B testing with real-world users, where it performed
significantly better than other systems. The results highlight the potential of
coupling ensemble systems with deep reinforcement learning as a fruitful path
for developing real-world, open-domain conversational agents.</description><identifier>DOI: 10.48550/arxiv.1801.06700</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Computation and Language ; Computer Science - Learning ; Computer Science - Neural and Evolutionary Computing ; Statistics - Machine Learning</subject><creationdate>2018-01</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/1801.06700$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1801.06700$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Serban, Iulian V</creatorcontrib><creatorcontrib>Sankar, Chinnadhurai</creatorcontrib><creatorcontrib>Germain, Mathieu</creatorcontrib><creatorcontrib>Zhang, Saizheng</creatorcontrib><creatorcontrib>Lin, Zhouhan</creatorcontrib><creatorcontrib>Subramanian, Sandeep</creatorcontrib><creatorcontrib>Kim, Taesup</creatorcontrib><creatorcontrib>Pieper, Michael</creatorcontrib><creatorcontrib>Chandar, Sarath</creatorcontrib><creatorcontrib>Ke, Nan Rosemary</creatorcontrib><creatorcontrib>Rajeswar, Sai</creatorcontrib><creatorcontrib>de Brebisson, Alexandre</creatorcontrib><creatorcontrib>Sotelo, Jose M. R</creatorcontrib><creatorcontrib>Suhubdy, Dendi</creatorcontrib><creatorcontrib>Michalski, Vincent</creatorcontrib><creatorcontrib>Nguyen, Alexandre</creatorcontrib><creatorcontrib>Pineau, Joelle</creatorcontrib><creatorcontrib>Bengio, Yoshua</creatorcontrib><title>A Deep Reinforcement Learning Chatbot (Short Version)</title><description>We present MILABOT: a deep reinforcement learning chatbot developed by the
Montreal Institute for Learning Algorithms (MILA) for the Amazon Alexa Prize
competition. MILABOT is capable of conversing with humans on popular small talk
topics through both speech and text. The system consists of an ensemble of
natural language generation and retrieval models, including neural network and
template-based models. By applying reinforcement learning to crowdsourced data
and real-world user interactions, the system has been trained to select an
appropriate response from the models in its ensemble. The system has been
evaluated through A/B testing with real-world users, where it performed
significantly better than other systems. The results highlight the potential of
coupling ensemble systems with deep reinforcement learning as a fruitful path
for developing real-world, open-domain conversational agents.</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Computation and Language</subject><subject>Computer Science - Learning</subject><subject>Computer Science - Neural and Evolutionary Computing</subject><subject>Statistics - Machine Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2018</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotzrFOwzAUQFEvDKjwAUx4hCHBzrNje6xCgUqRKtGqa_Qcv1BL1KncqIK_B0qnu10dxu6kKJXVWjxh_oqnUlohS1EbIa6ZnvNnogN_p5iGMfe0pzTxljCnmD54s8PJjxN_WO_GPPEt5WMc0-MNuxrw80i3l87Y5mWxad6KdvW6bOZtgb_3Ap0FA5J6Kaj2OjgzOGcqFawKQkuo0QfyWprgK9AAWLtBofO2VyFAZWHG7v-3Z3d3yHGP-bv783dnP_wA_ZM-kg</recordid><startdate>20180120</startdate><enddate>20180120</enddate><creator>Serban, Iulian V</creator><creator>Sankar, Chinnadhurai</creator><creator>Germain, Mathieu</creator><creator>Zhang, Saizheng</creator><creator>Lin, Zhouhan</creator><creator>Subramanian, Sandeep</creator><creator>Kim, Taesup</creator><creator>Pieper, Michael</creator><creator>Chandar, Sarath</creator><creator>Ke, Nan Rosemary</creator><creator>Rajeswar, Sai</creator><creator>de Brebisson, Alexandre</creator><creator>Sotelo, Jose M. R</creator><creator>Suhubdy, Dendi</creator><creator>Michalski, Vincent</creator><creator>Nguyen, Alexandre</creator><creator>Pineau, Joelle</creator><creator>Bengio, Yoshua</creator><scope>AKY</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20180120</creationdate><title>A Deep Reinforcement Learning Chatbot (Short Version)</title><author>Serban, Iulian V ; Sankar, Chinnadhurai ; Germain, Mathieu ; Zhang, Saizheng ; Lin, Zhouhan ; Subramanian, Sandeep ; Kim, Taesup ; Pieper, Michael ; Chandar, Sarath ; Ke, Nan Rosemary ; Rajeswar, Sai ; de Brebisson, Alexandre ; Sotelo, Jose M. R ; Suhubdy, Dendi ; Michalski, Vincent ; Nguyen, Alexandre ; Pineau, Joelle ; Bengio, Yoshua</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a670-a983731ec10e6b5d97f99724d84d05136abdeb517db23533a69f4a9b8c4dd3283</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Computation and Language</topic><topic>Computer Science - Learning</topic><topic>Computer Science - Neural and Evolutionary Computing</topic><topic>Statistics - Machine Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Serban, Iulian V</creatorcontrib><creatorcontrib>Sankar, Chinnadhurai</creatorcontrib><creatorcontrib>Germain, Mathieu</creatorcontrib><creatorcontrib>Zhang, Saizheng</creatorcontrib><creatorcontrib>Lin, Zhouhan</creatorcontrib><creatorcontrib>Subramanian, Sandeep</creatorcontrib><creatorcontrib>Kim, Taesup</creatorcontrib><creatorcontrib>Pieper, Michael</creatorcontrib><creatorcontrib>Chandar, Sarath</creatorcontrib><creatorcontrib>Ke, Nan Rosemary</creatorcontrib><creatorcontrib>Rajeswar, Sai</creatorcontrib><creatorcontrib>de Brebisson, Alexandre</creatorcontrib><creatorcontrib>Sotelo, Jose M. R</creatorcontrib><creatorcontrib>Suhubdy, Dendi</creatorcontrib><creatorcontrib>Michalski, Vincent</creatorcontrib><creatorcontrib>Nguyen, Alexandre</creatorcontrib><creatorcontrib>Pineau, Joelle</creatorcontrib><creatorcontrib>Bengio, Yoshua</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Serban, Iulian V</au><au>Sankar, Chinnadhurai</au><au>Germain, Mathieu</au><au>Zhang, Saizheng</au><au>Lin, Zhouhan</au><au>Subramanian, Sandeep</au><au>Kim, Taesup</au><au>Pieper, Michael</au><au>Chandar, Sarath</au><au>Ke, Nan Rosemary</au><au>Rajeswar, Sai</au><au>de Brebisson, Alexandre</au><au>Sotelo, Jose M. R</au><au>Suhubdy, Dendi</au><au>Michalski, Vincent</au><au>Nguyen, Alexandre</au><au>Pineau, Joelle</au><au>Bengio, Yoshua</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A Deep Reinforcement Learning Chatbot (Short Version)</atitle><date>2018-01-20</date><risdate>2018</risdate><abstract>We present MILABOT: a deep reinforcement learning chatbot developed by the
Montreal Institute for Learning Algorithms (MILA) for the Amazon Alexa Prize
competition. MILABOT is capable of conversing with humans on popular small talk
topics through both speech and text. The system consists of an ensemble of
natural language generation and retrieval models, including neural network and
template-based models. By applying reinforcement learning to crowdsourced data
and real-world user interactions, the system has been trained to select an
appropriate response from the models in its ensemble. The system has been
evaluated through A/B testing with real-world users, where it performed
significantly better than other systems. The results highlight the potential of
coupling ensemble systems with deep reinforcement learning as a fruitful path
for developing real-world, open-domain conversational agents.</abstract><doi>10.48550/arxiv.1801.06700</doi><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | DOI: 10.48550/arxiv.1801.06700 |
| ispartof | |
| issn | |
| language | eng |
| recordid | cdi_arxiv_primary_1801_06700 |
| source | arXiv.org |
| subjects | Computer Science - Artificial Intelligence Computer Science - Computation and Language Computer Science - Learning Computer Science - Neural and Evolutionary Computing Statistics - Machine Learning |
| title | A Deep Reinforcement Learning Chatbot (Short Version) |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-06T01%3A42%3A45IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20Deep%20Reinforcement%20Learning%20Chatbot%20(Short%20Version)&rft.au=Serban,%20Iulian%20V&rft.date=2018-01-20&rft_id=info:doi/10.48550/arxiv.1801.06700&rft_dat=%3Carxiv_GOX%3E1801_06700%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |