WEKO3
アイテム
{"_buckets": {"deposit": "69616dbd-5035-4931-95c8-44e03460e8e6"}, "_deposit": {"created_by": 29, "id": "1202", "owners": [29], "pid": {"revision_id": 0, "type": "depid", "value": "1202"}, "status": "published"}, "_oai": {"id": "oai:oist.repo.nii.ac.jp:00001202", "sets": ["78"]}, "author_link": ["6648", "6647"], "item_10001_biblio_info_7": {"attribute_name": "Bibliographic Information", "attribute_value_mlt": [{"bibliographicIssueDates": {"bibliographicIssueDate": "2017-12-15", "bibliographicIssueDateType": "Issued"}, "bibliographicIssueNumber": "1", "bibliographicPageEnd": "270", "bibliographicPageStart": "237", "bibliographicVolumeNumber": "30", "bibliographic_titles": [{}, {"bibliographic_title": "Neural Computation", "bibliographic_titleLang": "en"}]}]}, "item_10001_creator_3": {"attribute_name": "Author", "attribute_type": "creator", "attribute_value_mlt": [{"creatorNames": [{"creatorName": "Choi, Minkyu"}], "nameIdentifiers": [{"nameIdentifier": "6647", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Tani, Jun"}], "nameIdentifiers": [{"nameIdentifier": "6648", "nameIdentifierScheme": "WEKO"}]}]}, "item_10001_description_5": {"attribute_name": "Abstract", "attribute_value_mlt": [{"subitem_description": "This letter proposes a novel predictive coding type neural network model, the predictive multiple spatiotemporal scales recurrent neural network (P-MSTRNN). The P-MSTRNN learns to predict visually perceived human whole-body cyclic movement patterns by exploiting multiscale spatiotemporal constraints imposed on network dynamics by using differently sized receptive fields as well as different time constant values for each layer. After learning, the network can imitate target movement patterns by inferring or recognizing corresponding intentions by means of the regression of prediction error. Results show that the network can develop a functional hierarchy by developing a different type of dynamic structure at each layer. The letter examines how model performance during pattern generation, as well as predictive imitation, varies depending on the stage of learning. The number of limit cycle attractors corresponding to target movement patterns increases as learning proceeds. Transient dynamics developing early in the learning process successfully perform pattern generation and predictive imitation tasks. The letter concludes that exploitation of transient dynamics facilitates successful task performance during early learning periods.", "subitem_description_type": "Other"}]}, "item_10001_publisher_8": {"attribute_name": "Publisher", "attribute_value_mlt": [{"subitem_publisher": "The MIT Press Journals"}]}, "item_10001_relation_13": {"attribute_name": "PubMedNo.", "attribute_value_mlt": [{"subitem_relation_type": "isIdenticalTo", "subitem_relation_type_id": {"subitem_relation_type_id_text": "info:pmid/29064785", "subitem_relation_type_select": "PMID"}}]}, "item_10001_relation_14": {"attribute_name": "DOI", "attribute_value_mlt": [{"subitem_relation_type": "isIdenticalTo", "subitem_relation_type_id": {"subitem_relation_type_id_text": "info:doi/10.1162/neco_a_01026", "subitem_relation_type_select": "DOI"}}]}, "item_10001_relation_17": {"attribute_name": "Related site", "attribute_value_mlt": [{"subitem_relation_type_id": {"subitem_relation_type_id_text": "https://www.mitpressjournals.org/doi/abs/10.1162/neco_a_01026", "subitem_relation_type_select": "URI"}}]}, "item_10001_rights_15": {"attribute_name": "Rights", "attribute_value_mlt": [{"subitem_rights": "© 2017 Massachusetts Institute of Technology"}]}, "item_10001_source_id_9": {"attribute_name": "ISSN", "attribute_value_mlt": [{"subitem_source_identifier": "0899-7667", "subitem_source_identifier_type": "ISSN"}, {"subitem_source_identifier": "1530-888X", "subitem_source_identifier_type": "ISSN"}]}, "item_10001_version_type_20": {"attribute_name": "Author\u0027s flag", "attribute_value_mlt": [{"subitem_version_resource": "http://purl.org/coar/version/c_970fb48d4fbd8a85", "subitem_version_type": "VoR"}]}, "item_files": {"attribute_name": "ファイル情報", "attribute_type": "file", "attribute_value_mlt": [{"accessrole": "open_date", "date": [{"dateType": "Available", "dateValue": "2018-03-15"}], "displaytype": "detail", "download_preview_message": "", "file_order": 0, "filename": "Choi-2018-Predictive Coding for Dynamic Visual.pdf", "filesize": [{"value": "829.8 kB"}], "format": "application/pdf", "future_date_message": "", "is_thumbnail": false, "licensetype": "license_free", "mimetype": "application/pdf", "size": 829800.0, "url": {"label": "Choi-2018-Predictive Coding for Dynamic Visual", "url": "https://oist.repo.nii.ac.jp/record/1202/files/Choi-2018-Predictive Coding for Dynamic Visual.pdf"}, "version_id": "c3ebf64f-fff5-4999-bc02-419fd6af5648"}]}, "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": "journal article", "resourceuri": "http://purl.org/coar/resource_type/c_6501"}]}, "item_title": "Predictive Coding for Dynamic Visual Processing: Development of Functional Hierarchy in a Multiple Spatiotemporal Scales RNN Model", "item_titles": {"attribute_name": "タイトル", "attribute_value_mlt": [{"subitem_title": "Predictive Coding for Dynamic Visual Processing: Development of Functional Hierarchy in a Multiple Spatiotemporal Scales RNN Model", "subitem_title_language": "en"}]}, "item_type_id": "10001", "owner": "29", "path": ["78"], "permalink_uri": "https://oist.repo.nii.ac.jp/records/1202", "pubdate": {"attribute_name": "公開日", "attribute_value": "2020-02-26"}, "publish_date": "2020-02-26", "publish_status": "0", "recid": "1202", "relation": {}, "relation_version_is_last": true, "title": ["Predictive Coding for Dynamic Visual Processing: Development of Functional Hierarchy in a Multiple Spatiotemporal Scales RNN Model"], "weko_shared_id": 29}
Predictive Coding for Dynamic Visual Processing: Development of Functional Hierarchy in a Multiple Spatiotemporal Scales RNN Model
https://oist.repo.nii.ac.jp/records/1202
https://oist.repo.nii.ac.jp/records/1202e52e2c5d-8491-4992-abbb-0171e3a1a5d0
名前 / ファイル | ライセンス | アクション |
---|---|---|
Choi-2018-Predictive Coding for Dynamic Visual (829.8 kB)
|
|
Item type | 学術雑誌論文 / Journal Article(1) | |||||
---|---|---|---|---|---|---|
公開日 | 2020-02-26 | |||||
タイトル | ||||||
言語 | en | |||||
タイトル | Predictive Coding for Dynamic Visual Processing: Development of Functional Hierarchy in a Multiple Spatiotemporal Scales RNN Model | |||||
言語 | ||||||
言語 | eng | |||||
資源タイプ | ||||||
資源タイプ識別子 | http://purl.org/coar/resource_type/c_6501 | |||||
資源タイプ | journal article | |||||
著者(英) |
Choi, Minkyu
× Choi, Minkyu× Tani, Jun |
|||||
書誌情報 |
en : Neural Computation 巻 30, 号 1, p. 237-270, 発行日 2017-12-15 |
|||||
抄録 | ||||||
内容記述タイプ | Other | |||||
内容記述 | This letter proposes a novel predictive coding type neural network model, the predictive multiple spatiotemporal scales recurrent neural network (P-MSTRNN). The P-MSTRNN learns to predict visually perceived human whole-body cyclic movement patterns by exploiting multiscale spatiotemporal constraints imposed on network dynamics by using differently sized receptive fields as well as different time constant values for each layer. After learning, the network can imitate target movement patterns by inferring or recognizing corresponding intentions by means of the regression of prediction error. Results show that the network can develop a functional hierarchy by developing a different type of dynamic structure at each layer. The letter examines how model performance during pattern generation, as well as predictive imitation, varies depending on the stage of learning. The number of limit cycle attractors corresponding to target movement patterns increases as learning proceeds. Transient dynamics developing early in the learning process successfully perform pattern generation and predictive imitation tasks. The letter concludes that exploitation of transient dynamics facilitates successful task performance during early learning periods. | |||||
出版者 | ||||||
出版者 | The MIT Press Journals | |||||
ISSN | ||||||
収録物識別子タイプ | ISSN | |||||
収録物識別子 | 0899-7667 | |||||
ISSN | ||||||
収録物識別子タイプ | ISSN | |||||
収録物識別子 | 1530-888X | |||||
PubMed番号 | ||||||
関連タイプ | isIdenticalTo | |||||
識別子タイプ | PMID | |||||
関連識別子 | info:pmid/29064785 | |||||
DOI | ||||||
関連タイプ | isIdenticalTo | |||||
識別子タイプ | DOI | |||||
関連識別子 | info:doi/10.1162/neco_a_01026 | |||||
権利 | ||||||
権利情報 | © 2017 Massachusetts Institute of Technology | |||||
関連サイト | ||||||
識別子タイプ | URI | |||||
関連識別子 | https://www.mitpressjournals.org/doi/abs/10.1162/neco_a_01026 | |||||
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 |