WEKO3
アイテム
{"_buckets": {"deposit": "d4b2eec7-5cc4-4059-9261-7d23988ad669"}, "_deposit": {"created_by": 29, "id": "1736", "owners": [29], "pid": {"revision_id": 0, "type": "depid", "value": "1736"}, "status": "published"}, "_oai": {"id": "oai:oist.repo.nii.ac.jp:00001736", "sets": ["203", "222"]}, "author_link": ["1334", "8640", "479"], "item_10003_biblio_info_7": {"attribute_name": "書誌情報", "attribute_value_mlt": [{"bibliographicIssueDates": {"bibliographicIssueDate": "2019-09-26", "bibliographicIssueDateType": "Issued"}, "bibliographic_titles": [{}]}]}, "item_10003_description_5": {"attribute_name": "抄録", "attribute_value_mlt": [{"subitem_description": "In partially observable (PO) environments, deep reinforcement learning (RL) agents often suffer from unsatisfactory performance, since two problems need to be tackled together: how to extract information from the raw observations to solve the task, and how to improve the policy. In this study, we propose an RL algorithm for solving PO tasks. Our method comprises two parts: a variational recurrent model (VRM) for modeling the environment, and an RL controller that has access to both the environment and the VRM. The proposed algorithm was tested in two types of PO robotic control tasks, those in which either coordinates or velocities were not observable and those that require long-term memorization. Our experiments show that the proposed algorithm achieved better data efficiency and/or learned more optimal policy than other alternative approaches in tasks in which unobserved states cannot be inferred from raw observations in a simple manner.", "subitem_description_type": "Abstract"}]}, "item_10003_publisher_8": {"attribute_name": "出版者", "attribute_value_mlt": [{"subitem_publisher": "ICLR 2020"}]}, "item_10003_relation_17": {"attribute_name": "関連サイト", "attribute_value_mlt": [{"subitem_relation_name": [{"subitem_relation_name_text": "ICLR 2020"}], "subitem_relation_type_id": {"subitem_relation_type_id_text": "https://iclr.cc/Conferences/2020", "subitem_relation_type_select": "URI"}}, {"subitem_relation_type_id": {"subitem_relation_type_id_text": "https://iclr.cc/virtual/poster_r1lL4a4tDB.html", "subitem_relation_type_select": "URI"}}]}, "item_10003_rights_15": {"attribute_name": "権利", "attribute_value_mlt": [{"subitem_rights": "@ 2020 The Author(s)."}]}, "item_10003_version_type_20": {"attribute_name": "著者版フラグ", "attribute_value_mlt": [{"subitem_version_resource": "http://purl.org/coar/version/c_970fb48d4fbd8a85", "subitem_version_type": "VoR"}]}, "item_creator": {"attribute_name": "著者", "attribute_type": "creator", "attribute_value_mlt": [{"creatorNames": [{"creatorName": "Han, Dongqi", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "8640", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Doya, Kenji", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "479", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Tani, Jun", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "1334", "nameIdentifierScheme": "WEKO"}]}]}, "item_files": {"attribute_name": "ファイル情報", "attribute_type": "file", "attribute_value_mlt": [{"accessrole": "open_date", "date": [{"dateType": "Available", "dateValue": "2020-10-26"}], "displaytype": "detail", "download_preview_message": "", "file_order": 0, "filename": "variational_recurrent_models_for_solving_partially_observable_control_tasks.pdf", "filesize": [{"value": "9.8 MB"}], "format": "application/pdf", "future_date_message": "", "is_thumbnail": false, "licensetype": "license_free", "mimetype": "application/pdf", "size": 9800000.0, "url": {"label": "variational_recurrent_models_for_solving_partially_observable_control_tasks", "url": "https://oist.repo.nii.ac.jp/record/1736/files/variational_recurrent_models_for_solving_partially_observable_control_tasks.pdf"}, "version_id": "1ded4e1e-ec26-4a74-937f-727cd6b2c81e"}]}, "item_keyword": {"attribute_name": "キーワード", "attribute_value_mlt": [{"subitem_subject": "Reinforcement Learning", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}, {"subitem_subject": "Deep Learning", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}, {"subitem_subject": "Variational Inference", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}, {"subitem_subject": "Recurrent Neural Network", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}, {"subitem_subject": "Partially Observable", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}, {"subitem_subject": "Robotic Control", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}, {"subitem_subject": "Continuous Control", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}]}, "item_language": {"attribute_name": "言語", "attribute_value_mlt": [{"subitem_language": "eng"}]}, "item_resource_type": {"attribute_name": "資源タイプ", "attribute_value_mlt": [{"resourcetype": "conference paper", "resourceuri": "http://purl.org/coar/resource_type/c_5794"}]}, "item_title": "Variational Recurrent Models for Solving Partially Observable Control Tasks", "item_titles": {"attribute_name": "タイトル", "attribute_value_mlt": [{"subitem_title": "Variational Recurrent Models for Solving Partially Observable Control Tasks", "subitem_title_language": "en"}]}, "item_type_id": "10003", "owner": "29", "path": ["203", "222"], "permalink_uri": "https://oist.repo.nii.ac.jp/records/1736", "pubdate": {"attribute_name": "公開日", "attribute_value": "2020-10-26"}, "publish_date": "2020-10-26", "publish_status": "0", "recid": "1736", "relation": {}, "relation_version_is_last": true, "title": ["Variational Recurrent Models for Solving Partially Observable Control Tasks"], "weko_shared_id": 29}
Variational Recurrent Models for Solving Partially Observable Control Tasks
https://oist.repo.nii.ac.jp/records/1736
https://oist.repo.nii.ac.jp/records/1736e0c74953-5266-4f3f-81be-03cbdef0292c
名前 / ファイル | ライセンス | アクション |
---|---|---|
variational_recurrent_models_for_solving_partially_observable_control_tasks (9.8 MB)
|
|
Item type | 会議発表論文 / Conference Paper(1) | |||||
---|---|---|---|---|---|---|
公開日 | 2020-10-26 | |||||
タイトル | ||||||
言語 | en | |||||
タイトル | Variational Recurrent Models for Solving Partially Observable Control Tasks | |||||
言語 | ||||||
言語 | eng | |||||
キーワード | ||||||
言語 | en | |||||
主題Scheme | Other | |||||
主題 | Reinforcement Learning | |||||
キーワード | ||||||
言語 | en | |||||
主題Scheme | Other | |||||
主題 | Deep Learning | |||||
キーワード | ||||||
言語 | en | |||||
主題Scheme | Other | |||||
主題 | Variational Inference | |||||
キーワード | ||||||
言語 | en | |||||
主題Scheme | Other | |||||
主題 | Recurrent Neural Network | |||||
キーワード | ||||||
言語 | en | |||||
主題Scheme | Other | |||||
主題 | Partially Observable | |||||
キーワード | ||||||
言語 | en | |||||
主題Scheme | Other | |||||
主題 | Robotic Control | |||||
キーワード | ||||||
言語 | en | |||||
主題Scheme | Other | |||||
主題 | Continuous Control | |||||
資源タイプ | ||||||
資源タイプ識別子 | http://purl.org/coar/resource_type/c_5794 | |||||
資源タイプ | conference paper | |||||
著者 |
Han, Dongqi
× Han, Dongqi× Doya, Kenji× Tani, Jun |
|||||
書誌情報 | 発行日 2019-09-26 | |||||
抄録 | ||||||
内容記述タイプ | Abstract | |||||
内容記述 | In partially observable (PO) environments, deep reinforcement learning (RL) agents often suffer from unsatisfactory performance, since two problems need to be tackled together: how to extract information from the raw observations to solve the task, and how to improve the policy. In this study, we propose an RL algorithm for solving PO tasks. Our method comprises two parts: a variational recurrent model (VRM) for modeling the environment, and an RL controller that has access to both the environment and the VRM. The proposed algorithm was tested in two types of PO robotic control tasks, those in which either coordinates or velocities were not observable and those that require long-term memorization. Our experiments show that the proposed algorithm achieved better data efficiency and/or learned more optimal policy than other alternative approaches in tasks in which unobserved states cannot be inferred from raw observations in a simple manner. | |||||
出版者 | ||||||
出版者 | ICLR 2020 | |||||
権利 | ||||||
権利情報 | @ 2020 The Author(s). | |||||
関連サイト | ||||||
識別子タイプ | URI | |||||
関連識別子 | https://iclr.cc/Conferences/2020 | |||||
関連名称 | ICLR 2020 | |||||
関連サイト | ||||||
識別子タイプ | URI | |||||
関連識別子 | https://iclr.cc/virtual/poster_r1lL4a4tDB.html | |||||
著者版フラグ | ||||||
出版タイプ | VoR | |||||
出版タイプResource | http://purl.org/coar/version/c_970fb48d4fbd8a85 |