WEKO3
インデックスリンク
アイテム
{"_buckets": {"deposit": "9ea992a9-5630-4e3c-a848-8d5c16a81487"}, "_deposit": {"created_by": 3, "id": "7967", "owners": [3], "pid": {"revision_id": 0, "type": "depid", "value": "7967"}, "status": "published"}, "_oai": {"id": "oai:fukuyama-u.repo.nii.ac.jp:00007967", "sets": ["700"]}, "author_link": ["42812", "42813", "42811"], "item_1_biblio_info_14": {"attribute_name": "書誌情報", "attribute_value_mlt": [{"bibliographicIssueDates": {"bibliographicIssueDate": "1996-03", "bibliographicIssueDateType": "Issued"}, "bibliographicIssueNumber": "2", "bibliographicPageEnd": "44", "bibliographicPageStart": "37", "bibliographicVolumeNumber": "19", "bibliographic_titles": [{"bibliographic_title": "福山大学工学部紀要"}]}]}, "item_1_creator_6": {"attribute_name": "著者名(日)", "attribute_type": "creator", "attribute_value_mlt": [{"creatorNames": [{"creatorName": "渡辺, 栄治"}], "nameIdentifiers": [{"nameIdentifier": "42811", "nameIdentifierScheme": "WEKO"}]}]}, "item_1_description_1": {"attribute_name": "ページ属性", "attribute_value_mlt": [{"subitem_description": "P(論文)", "subitem_description_type": "Other"}]}, "item_1_description_12": {"attribute_name": "抄録(英)", "attribute_value_mlt": [{"subitem_description": "This paper discusses the relation between internal representations and generalization ability of multi-layered neural networks for function approximation problems. Here, the over learning problem, which makes the generalization ability poor, is divided into the two problems; the excessive degrees of freedom and the non-uniqueness of weights. These problems are carefully discussed from the viewpoint of internal representation; weights and outputs of hidden units. First, the excessive degrees of freedom problem is discussed by introducing the entropy for weights. Next, the non-uniqueness of weights problem is also discussed by introducing the principal components analysis method. Finally, the relations between internal representations and generalization ability of multi-layered neural networks are concretely discussed by numerical results.", "subitem_description_type": "Other"}]}, "item_1_full_name_7": {"attribute_name": "著者名よみ", "attribute_value_mlt": [{"nameIdentifiers": [{"nameIdentifier": "42812", "nameIdentifierScheme": "WEKO"}], "names": [{"name": "ワタナベ, エイジ"}]}]}, "item_1_full_name_8": {"attribute_name": "著者名(英)", "attribute_value_mlt": [{"nameIdentifiers": [{"nameIdentifier": "42813", "nameIdentifierScheme": "WEKO"}], "names": [{"name": "WATANABE, Eiji", "nameLang": "en"}]}]}, "item_1_source_id_13": {"attribute_name": "雑誌書誌ID", "attribute_value_mlt": [{"subitem_source_identifier": "AN00217655", "subitem_source_identifier_type": "NCID"}]}, "item_1_text_9": {"attribute_name": "著者所属(日)", "attribute_value_mlt": [{"subitem_text_value": "福山大学工学部情報処理工学科"}]}, "item_files": {"attribute_name": "ファイル情報", "attribute_type": "file", "attribute_value_mlt": [{"accessrole": "open_date", "date": [{"dateType": "Available", "dateValue": "1996-03-01"}], "displaytype": "detail", "download_preview_message": "", "file_order": 0, "filename": "KJ00005781366.pdf", "filesize": [{"value": "606.9 kB"}], "format": "application/pdf", "future_date_message": "", "is_thumbnail": false, "licensetype": "license_free", "mimetype": "application/pdf", "size": 606900.0, "url": {"url": "https://fukuyama-u.repo.nii.ac.jp/record/7967/files/KJ00005781366.pdf"}, "version_id": "6c5df41b-b814-4eae-a6ec-4a7466960cd7"}]}, "item_keyword": {"attribute_name": "キーワード", "attribute_value_mlt": [{"subitem_subject": "階層型ニューラルネットワーク", "subitem_subject_scheme": "Other"}, {"subitem_subject": "関数近似問題", "subitem_subject_scheme": "Other"}, {"subitem_subject": "内部表現", "subitem_subject_scheme": "Other"}, {"subitem_subject": "汎化能力", "subitem_subject_scheme": "Other"}, {"subitem_subject": "過剰学習", "subitem_subject_scheme": "Other"}, {"subitem_subject": "Multi-layered neural network", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}, {"subitem_subject": "function approximation problem", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}, {"subitem_subject": "internal representation", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}, {"subitem_subject": "generalization ability", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}, {"subitem_subject": "over learning", "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": "関数近似問題に対する階層型ニューラルネットワークの内部表現と汎化能力の関係", "item_titles": {"attribute_name": "タイトル", "attribute_value_mlt": [{"subitem_title": "関数近似問題に対する階層型ニューラルネットワークの内部表現と汎化能力の関係"}, {"subitem_title": "Relations between Internal Representations and Generalization Ability of Multi Layered Neural Networks for Function Approximation Problems", "subitem_title_language": "en"}]}, "item_type_id": "1", "owner": "3", "path": ["700"], "permalink_uri": "https://fukuyama-u.repo.nii.ac.jp/records/7967", "pubdate": {"attribute_name": "公開日", "attribute_value": "1996-03-01"}, "publish_date": "1996-03-01", "publish_status": "0", "recid": "7967", "relation": {}, "relation_version_is_last": true, "title": ["関数近似問題に対する階層型ニューラルネットワークの内部表現と汎化能力の関係"], "weko_shared_id": -1}
関数近似問題に対する階層型ニューラルネットワークの内部表現と汎化能力の関係
https://fukuyama-u.repo.nii.ac.jp/records/7967
https://fukuyama-u.repo.nii.ac.jp/records/796774d671f0-62f5-4ef2-9230-317caff87de2
名前 / ファイル | ライセンス | アクション |
---|---|---|
KJ00005781366.pdf (606.9 kB)
|
|
Item type | 紀要論文(ELS) / Departmental Bulletin Paper(1) | |||||
---|---|---|---|---|---|---|
公開日 | 1996-03-01 | |||||
タイトル | ||||||
タイトル | 関数近似問題に対する階層型ニューラルネットワークの内部表現と汎化能力の関係 | |||||
タイトル | ||||||
言語 | en | |||||
タイトル | Relations between Internal Representations and Generalization Ability of Multi Layered Neural Networks for Function Approximation Problems | |||||
言語 | ||||||
言語 | jpn | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | 階層型ニューラルネットワーク | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | 関数近似問題 | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | 内部表現 | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | 汎化能力 | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | 過剰学習 | |||||
キーワード | ||||||
言語 | en | |||||
主題Scheme | Other | |||||
主題 | Multi-layered neural network | |||||
キーワード | ||||||
言語 | en | |||||
主題Scheme | Other | |||||
主題 | function approximation problem | |||||
キーワード | ||||||
言語 | en | |||||
主題Scheme | Other | |||||
主題 | internal representation | |||||
キーワード | ||||||
言語 | en | |||||
主題Scheme | Other | |||||
主題 | generalization ability | |||||
キーワード | ||||||
言語 | en | |||||
主題Scheme | Other | |||||
主題 | over learning | |||||
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 | |||||
ページ属性 | ||||||
内容記述タイプ | Other | |||||
内容記述 | P(論文) | |||||
著者名(日) |
渡辺, 栄治
× 渡辺, 栄治 |
|||||
著者名よみ | ||||||
識別子Scheme | WEKO | |||||
識別子 | 42812 | |||||
姓名 | ワタナベ, エイジ | |||||
著者名(英) | ||||||
識別子Scheme | WEKO | |||||
識別子 | 42813 | |||||
姓名 | WATANABE, Eiji | |||||
言語 | en | |||||
著者所属(日) | ||||||
値 | 福山大学工学部情報処理工学科 | |||||
抄録(英) | ||||||
内容記述タイプ | Other | |||||
内容記述 | This paper discusses the relation between internal representations and generalization ability of multi-layered neural networks for function approximation problems. Here, the over learning problem, which makes the generalization ability poor, is divided into the two problems; the excessive degrees of freedom and the non-uniqueness of weights. These problems are carefully discussed from the viewpoint of internal representation; weights and outputs of hidden units. First, the excessive degrees of freedom problem is discussed by introducing the entropy for weights. Next, the non-uniqueness of weights problem is also discussed by introducing the principal components analysis method. Finally, the relations between internal representations and generalization ability of multi-layered neural networks are concretely discussed by numerical results. | |||||
雑誌書誌ID | ||||||
収録物識別子タイプ | NCID | |||||
収録物識別子 | AN00217655 | |||||
書誌情報 |
福山大学工学部紀要 巻 19, 号 2, p. 37-44, 発行日 1996-03 |