WEKO3
アイテム
{"_buckets": {"deposit": "54ac8972-6d9b-41ca-ace0-bbbf86661e32"}, "_deposit": {"created_by": 5, "id": "2701", "owner": "5", "owners": [5], "pid": {"revision_id": 0, "type": "depid", "value": "2701"}, "status": "published"}, "_oai": {"id": "oai:miyazaki-u.repo.nii.ac.jp:00002701", "sets": []}, "author_link": ["11805", "14111", "11807", "14114"], "item_10002_alternative_title_1": {"attribute_name": "その他(別言語等)のタイトル", "attribute_value_mlt": [{"subitem_alternative_title": "BPA-PRホウ ニ ヨル カイソウガタ ニューラルネットワーク ノ コショウ ホショウ ノウリョク ノ コウジョウ", "subitem_alternative_title_language": "ja-Kana"}]}, "item_10002_biblio_info_7": {"attribute_name": "書誌情報", "attribute_value_mlt": [{"bibliographicIssueDates": {"bibliographicIssueDate": "2008-08-30", "bibliographicIssueDateType": "Issued"}, "bibliographicPageEnd": "280", "bibliographicPageStart": "275", "bibliographicVolumeNumber": "37", "bibliographic_titles": [{"bibliographic_title": "宮崎大学工学部紀要", "bibliographic_titleLang": "ja"}, {"bibliographic_title": "Memoirs of Faculty of Engineering, University of Miyazaki", "bibliographic_titleLang": "en"}]}]}, "item_10002_description_5": {"attribute_name": "抄録", "attribute_value_mlt": [{"subitem_description": "Some researchers have been proposed to implement neural networks into Wafer Scale Integration(WSI) to achieve fast learning. When neural networks are implemented into a WSI, it has to have a mechanism to avoid hardware defects. To compensate hardware defects, the partial retraining (PR) scheme has proposed. The performance of PR scheme depends on the weights in the neural network because PR scheme only adjusts the weights belonging to a neuron affected by the defects. In this paper, we propose back propagation with annealing scheme (BPA scheme) to improve defect compensation ratio. We show that BPA scheme achieved higher capability of defect compensation than that of conventional BP algorithm.", "subitem_description_language": "en", "subitem_description_type": "Abstract"}]}, "item_10002_publisher_8": {"attribute_name": "出版者", "attribute_value_mlt": [{"subitem_publisher": "宮崎大学工学部", "subitem_publisher_language": "ja"}, {"subitem_publisher": "Faculty of Engineering, University of Miyazaki", "subitem_publisher_language": "en"}]}, "item_10002_source_id_11": {"attribute_name": "書誌レコードID", "attribute_value_mlt": [{"subitem_source_identifier": "AA00732558", "subitem_source_identifier_type": "NCID"}]}, "item_10002_source_id_9": {"attribute_name": "ISSN", "attribute_value_mlt": [{"subitem_source_identifier": "05404924", "subitem_source_identifier_type": "ISSN"}]}, "item_10002_text_25": {"attribute_name": "BIBID", "attribute_value_mlt": [{"subitem_text_value": "TC00002694"}]}, "item_10002_text_26": {"attribute_name": "登録日", "attribute_value_mlt": [{"subitem_text_value": "2008/11/11"}]}, "item_10002_textarea_24": {"attribute_name": "来歴", "attribute_value_mlt": [{"subitem_textarea_value": "Submitted by dspace dspace (lib-ir@of.miyazaki-u.ac.jp) on 2008-11-10T09:35:12Z\nNo. of bitstreams: 2\nlicence.pdf: 24474 bytes, checksum: ffda027e83e1c8d1daa69d4fd24957a0 (MD5)\nKJ00005016062.pdf: 324344 bytes, checksum: 6d37ea7d637242040fd6d6e5c2277d3a (MD5)"}, {"subitem_textarea_value": "Approved for entry into archive by dspace dspace(lib-ir@of.miyazaki-u.ac.jp) on 2008-11-11T04:34:42Z (GMT) No. of bitstreams: 2\nlicence.pdf: 24474 bytes, checksum: ffda027e83e1c8d1daa69d4fd24957a0 (MD5)\nKJ00005016062.pdf: 324344 bytes, checksum: 6d37ea7d637242040fd6d6e5c2277d3a (MD5)"}, {"subitem_textarea_value": "Made available in DSpace on 2008-11-11T04:34:42Z (GMT). No. of bitstreams: 2\nlicence.pdf: 24474 bytes, checksum: ffda027e83e1c8d1daa69d4fd24957a0 (MD5)\nKJ00005016062.pdf: 324344 bytes, checksum: 6d37ea7d637242040fd6d6e5c2277d3a (MD5)\n Previous issue date: 2008-08-30"}]}, "item_10002_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": "山森, 一人", "creatorNameLang": "ja"}, {"creatorName": "ヤマモリ, クニヒト", "creatorNameLang": "ja-Kana"}, {"creatorName": "Yamamori, Kunihito", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "11805", "nameIdentifierScheme": "WEKO"}, {"nameIdentifier": "50293395", "nameIdentifierScheme": "e-Rad", "nameIdentifierURI": "https://kaken.nii.ac.jp/ja/search/?qm=50293395"}]}, {"creatorNames": [{"creatorName": "森元, 聡明"}, {"creatorName": "モリモト, アキラ", "creatorNameLang": "ja-Kana"}], "nameIdentifiers": [{"nameIdentifier": "14111", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "吉原, 郁夫", "creatorNameLang": "ja"}, {"creatorName": "ヨシハラ, イクオ", "creatorNameLang": "ja-Kana"}, {"creatorName": "Yoshihara, Ikuo", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "11807", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Morimoto, Akira", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "14114", "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": "2020-06-21"}], "displaytype": "detail", "download_preview_message": "", "file_order": 0, "filename": "KJ00005016062.pdf", "filesize": [{"value": "324.3 kB"}], "format": "application/pdf", "future_date_message": "", "is_thumbnail": false, "licensetype": "license_free", "mimetype": "application/pdf", "size": 324300.0, "url": {"label": "KJ00005016062.pdf", "url": "https://miyazaki-u.repo.nii.ac.jp/record/2701/files/KJ00005016062.pdf"}, "version_id": "e658594e-b5ff-4815-9b78-b9482638e8e7"}]}, "item_keyword": {"attribute_name": "キーワード", "attribute_value_mlt": [{"subitem_subject": "Back Propagation with annealing, Partial Retraining scheme, Multi-Layer Neural Network, Fault Tolerance", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}]}, "item_language": {"attribute_name": "言語", "attribute_value_mlt": [{"subitem_language": "jpn"}]}, "item_resource_type": {"attribute_name": "資源タイプ", "attribute_value_mlt": [{"resourcetype": "departmental bulletin paper", "resourceuri": "http://purl.org/coar/resource_type/c_6501"}]}, "item_title": "BPA-PR法による階層型ニューラルネットワークの故障補償能力の向上", "item_titles": {"attribute_name": "タイトル", "attribute_value_mlt": [{"subitem_title": "BPA-PR法による階層型ニューラルネットワークの故障補償能力の向上", "subitem_title_language": "ja"}, {"subitem_title": "Defect Compensation method for Multi-Layer Neural Network by PR Scheme with BPA algorithm", "subitem_title_language": "en"}]}, "item_type_id": "10002", "owner": "5", "path": ["73", "36", "330", "318"], "permalink_uri": "http://hdl.handle.net/10458/1637", "pubdate": {"attribute_name": "PubDate", "attribute_value": "2008-11-11"}, "publish_date": "2008-11-11", "publish_status": "0", "recid": "2701", "relation": {}, "relation_version_is_last": true, "title": ["BPA-PR法による階層型ニューラルネットワークの故障補償能力の向上"], "weko_shared_id": 2}
BPA-PR法による階層型ニューラルネットワークの故障補償能力の向上
http://hdl.handle.net/10458/1637
http://hdl.handle.net/10458/16377aeb084b-4026-4d0d-9813-31498716155c
名前 / ファイル | ライセンス | アクション |
---|---|---|
KJ00005016062.pdf (324.3 kB)
|
|
Item type | 紀要論文 / Departmental Bulletin Paper(1) | |||||
---|---|---|---|---|---|---|
公開日 | 2008-11-11 | |||||
タイトル | ||||||
言語 | ja | |||||
タイトル | BPA-PR法による階層型ニューラルネットワークの故障補償能力の向上 | |||||
タイトル | ||||||
言語 | en | |||||
タイトル | Defect Compensation method for Multi-Layer Neural Network by PR Scheme with BPA algorithm | |||||
言語 | ||||||
言語 | jpn | |||||
キーワード | ||||||
言語 | en | |||||
主題Scheme | Other | |||||
主題 | Back Propagation with annealing, Partial Retraining scheme, Multi-Layer Neural Network, Fault Tolerance | |||||
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 | departmental bulletin paper | |||||
その他(別言語等)のタイトル | ||||||
その他のタイトル | BPA-PRホウ ニ ヨル カイソウガタ ニューラルネットワーク ノ コショウ ホショウ ノウリョク ノ コウジョウ | |||||
言語 | ja-Kana | |||||
著者 |
山森, 一人
× 山森, 一人× 森元, 聡明× 吉原, 郁夫× Morimoto, Akira |
|||||
抄録 | ||||||
内容記述タイプ | Abstract | |||||
内容記述 | Some researchers have been proposed to implement neural networks into Wafer Scale Integration(WSI) to achieve fast learning. When neural networks are implemented into a WSI, it has to have a mechanism to avoid hardware defects. To compensate hardware defects, the partial retraining (PR) scheme has proposed. The performance of PR scheme depends on the weights in the neural network because PR scheme only adjusts the weights belonging to a neuron affected by the defects. In this paper, we propose back propagation with annealing scheme (BPA scheme) to improve defect compensation ratio. We show that BPA scheme achieved higher capability of defect compensation than that of conventional BP algorithm. | |||||
言語 | en | |||||
書誌情報 |
ja : 宮崎大学工学部紀要 en : Memoirs of Faculty of Engineering, University of Miyazaki 巻 37, p. 275-280, 発行日 2008-08-30 |
|||||
出版者 | ||||||
言語 | ja | |||||
出版者 | 宮崎大学工学部 | |||||
出版者 | ||||||
言語 | en | |||||
出版者 | Faculty of Engineering, University of Miyazaki | |||||
ISSN | ||||||
収録物識別子タイプ | ISSN | |||||
収録物識別子 | 05404924 | |||||
書誌レコードID | ||||||
収録物識別子タイプ | NCID | |||||
収録物識別子 | AA00732558 | |||||
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 |