WEKO3
アイテム
{"_buckets": {"deposit": "e3288138-5555-4f54-8971-8e3392e38534"}, "_deposit": {"created_by": 3, "id": "1634", "owners": [3], "pid": {"revision_id": 0, "type": "depid", "value": "1634"}, "status": "published"}, "_oai": {"id": "oai:kyutech.repo.nii.ac.jp:00001634", "sets": ["9"]}, "author_link": ["5343", "6608", "6609", "6610"], "control_number": "1634", "item_21_biblio_info_6": {"attribute_name": "書誌情報", "attribute_value_mlt": [{"bibliographicIssueDates": {"bibliographicIssueDate": "2008-04", "bibliographicIssueDateType": "Issued"}, "bibliographicIssueNumber": "2", "bibliographicPageEnd": "34", "bibliographicPageStart": "27", "bibliographicVolumeNumber": "13", "bibliographic_titles": [{"bibliographic_title": "International Journal of Biomedical Soft Computing and Human Sciences"}]}]}, "item_21_description_4": {"attribute_name": "抄録", "attribute_value_mlt": [{"subitem_description": "We propose a new associative memory to improve its noise tolerance and storage capacity. Our underlying model is an improved multidirectional associative memory (IMAM), which uses autoassociative bottleneck neural networks to remove noise in its input, i.e., analyze patterns. IMAM has inefficient storage capacity and low noise tolerance due to a correlation matrix representing association. One of our basic ideas is to replace a correlation matrix with a multilayer perceptron (MLP), which has better learning and generalization capability. Moreover, we introduce two improvements. One is to add intermediate elements into MLP to improve its performance. The other is to use outputs of hidden layers in a five-layer bottleneck neural network. These outputs include information on synthesis of a key pattern from compressed information in the middle layer. To evaluate the proposed approaches, we compared three types of associative memory: associative memory with a bottleneck neural network and MLP (AM/B-M), AM/B-M with intermediate elements (AM/B-I), and AM/B-I with synthetic outputs (AM/B-IS). 10-by-10 images of Latin alphabet are used as patterns for association. In a case of association between 78 non-injective pattern pairs with 10% noise, our proposed AM/B-IS is better than AM/B-M by more than 40% in pattern recalling ratio.", "subitem_description_type": "Abstract"}]}, "item_21_description_60": {"attribute_name": "資源タイプ", "attribute_value_mlt": [{"subitem_description": "Journal Article", "subitem_description_type": "Other"}]}, "item_21_full_name_3": {"attribute_name": "著者別名", "attribute_value_mlt": [{"nameIdentifiers": [{"nameIdentifier": "6610", "nameIdentifierScheme": "WEKO"}], "names": [{"name": "横井, 博一", "nameLang": "ja"}]}]}, "item_21_publisher_7": {"attribute_name": "出版者", "attribute_value_mlt": [{"subitem_publisher": "バイオメディカル・ファジィ・システム学会"}]}, "item_21_relation_66": {"attribute_name": "論文ID(NAID)", "attribute_value_mlt": [{"subitem_relation_type": "isIdenticalTo", "subitem_relation_type_id": {"subitem_relation_type_id_text": "110006991259", "subitem_relation_type_select": "NAID"}}]}, "item_21_rights_13": {"attribute_name": "権利", "attribute_value_mlt": [{"subitem_rights": "Copyright©1995 Biomedical Fuzzy Systems Association. 本文データは学協会の許諾に基づきCiNiiから複製したものである"}]}, "item_21_select_59": {"attribute_name": "査読の有無", "attribute_value_mlt": [{"subitem_select_item": "yes"}]}, "item_21_source_id_8": {"attribute_name": "ISSN", "attribute_value_mlt": [{"subitem_source_identifier": "1345-1529", "subitem_source_identifier_type": "PISSN"}]}, "item_21_text_36": {"attribute_name": "著者所属", "attribute_value_mlt": [{"subitem_text_value": "Kyushu Institute of Technology"}, {"subitem_text_value": "Tohoku University"}, {"subitem_text_value": "Kyushu Institute of Technology"}]}, "item_21_text_64": {"attribute_name": "業績ID", "attribute_value_mlt": [{"subitem_text_value": "6E92DB0348374B3B4925762B001D34B5"}]}, "item_21_version_type_58": {"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": [{"creatorAffiliations": [{"affiliationNameIdentifiers": [], "affiliationNames": [{"affiliationName": "", "affiliationNameLang": "ja"}]}], "creatorNames": [{"creatorName": "Inohira, Eiichi", "creatorNameLang": "en"}, {"creatorName": "猪平, 栄一", "creatorNameLang": "ja"}, {"creatorName": "イノヒラ, エイイチ", "creatorNameLang": "ja-Kana"}], "familyNames": [{"familyName": "Inohira", "familyNameLang": "en"}, {"familyName": "猪平", "familyNameLang": "ja"}, {"familyName": "イノヒラ", "familyNameLang": "ja-Kana"}], "givenNames": [{"givenName": "Eiichi", "givenNameLang": "en"}, {"givenName": "栄一", "givenNameLang": "ja"}, {"givenName": "エイイチ", "givenNameLang": "ja-Kana"}], "nameIdentifiers": [{"nameIdentifier": "5343", "nameIdentifierScheme": "WEKO"}, {"nameIdentifier": "70363405", "nameIdentifierScheme": "e-Rad", "nameIdentifierURI": "https://nrid.nii.ac.jp/ja/nrid/1000070363405"}, {"nameIdentifier": "6508016934", "nameIdentifierScheme": "Scopus著者ID", "nameIdentifierURI": "https://www.scopus.com/authid/detail.uri?authorId=6508016934"}, {"nameIdentifier": "329", "nameIdentifierScheme": "九工大研究者情報", "nameIdentifierURI": "https://hyokadb02.jimu.kyutech.ac.jp/html/329_ja.html"}]}, {"creatorNames": [{"creatorName": "Ogawa, Takeshi", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "6608", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Yokoi, Hirokazu", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "6609", "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": "2009-09-09"}], "displaytype": "detail", "download_preview_message": "", "file_order": 0, "filename": "Associative memory.pdf", "filesize": [{"value": "842.8 kB"}], "format": "application/pdf", "future_date_message": "", "is_thumbnail": false, "licensetype": "license_free", "mimetype": "application/pdf", "size": 842800.0, "url": {"label": "Associative memory.pdf", "url": "https://kyutech.repo.nii.ac.jp/record/1634/files/Associative memory.pdf"}, "version_id": "cf16abac-8783-4574-b242-7d69e0413b03"}]}, "item_keyword": {"attribute_name": "キーワード", "attribute_value_mlt": [{"subitem_subject": "Associative memory", "subitem_subject_scheme": "Other"}, {"subitem_subject": "bottleneck neural network", "subitem_subject_scheme": "Other"}, {"subitem_subject": "multilayer perceptron", "subitem_subject_scheme": "Other"}, {"subitem_subject": "intermediate element", "subitem_subject_scheme": "Other"}, {"subitem_subject": "noise tolerance", "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": "journal article", "resourceuri": "http://purl.org/coar/resource_type/c_6501"}]}, "item_title": "Associative Memory with Pattern Analysis and Synthesis by a Bottleneck Neural Network", "item_titles": {"attribute_name": "タイトル", "attribute_value_mlt": [{"subitem_title": "Associative Memory with Pattern Analysis and Synthesis by a Bottleneck Neural Network", "subitem_title_language": "en"}]}, "item_type_id": "21", "owner": "3", "path": ["9"], "permalink_uri": "http://hdl.handle.net/10228/2478", "pubdate": {"attribute_name": "PubDate", "attribute_value": "2009-09-09"}, "publish_date": "2009-09-09", "publish_status": "0", "recid": "1634", "relation": {}, "relation_version_is_last": true, "title": ["Associative Memory with Pattern Analysis and Synthesis by a Bottleneck Neural Network"], "weko_shared_id": -1}
Associative Memory with Pattern Analysis and Synthesis by a Bottleneck Neural Network
http://hdl.handle.net/10228/2478
http://hdl.handle.net/10228/2478d3fe361d-8d6b-4e6d-ac95-afe6d889c725
名前 / ファイル | ライセンス | アクション |
---|---|---|
Associative memory.pdf (842.8 kB)
|
|
Item type | 学術雑誌論文 = Journal Article(1) | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
公開日 | 2009-09-09 | |||||||||||
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 | |||||||||||
タイトル | ||||||||||||
言語 | en | |||||||||||
タイトル | Associative Memory with Pattern Analysis and Synthesis by a Bottleneck Neural Network | |||||||||||
言語 | ||||||||||||
言語 | eng | |||||||||||
著者 |
猪平, 栄一
× 猪平, 栄一
WEKO
5343
× Ogawa, Takeshi× Yokoi, Hirokazu |
|||||||||||
抄録 | ||||||||||||
内容記述タイプ | Abstract | |||||||||||
内容記述 | We propose a new associative memory to improve its noise tolerance and storage capacity. Our underlying model is an improved multidirectional associative memory (IMAM), which uses autoassociative bottleneck neural networks to remove noise in its input, i.e., analyze patterns. IMAM has inefficient storage capacity and low noise tolerance due to a correlation matrix representing association. One of our basic ideas is to replace a correlation matrix with a multilayer perceptron (MLP), which has better learning and generalization capability. Moreover, we introduce two improvements. One is to add intermediate elements into MLP to improve its performance. The other is to use outputs of hidden layers in a five-layer bottleneck neural network. These outputs include information on synthesis of a key pattern from compressed information in the middle layer. To evaluate the proposed approaches, we compared three types of associative memory: associative memory with a bottleneck neural network and MLP (AM/B-M), AM/B-M with intermediate elements (AM/B-I), and AM/B-I with synthetic outputs (AM/B-IS). 10-by-10 images of Latin alphabet are used as patterns for association. In a case of association between 78 non-injective pattern pairs with 10% noise, our proposed AM/B-IS is better than AM/B-M by more than 40% in pattern recalling ratio. | |||||||||||
書誌情報 |
International Journal of Biomedical Soft Computing and Human Sciences 巻 13, 号 2, p. 27-34, 発行日 2008-04 |
|||||||||||
出版社 | ||||||||||||
出版者 | バイオメディカル・ファジィ・システム学会 | |||||||||||
論文ID(NAID) | ||||||||||||
関連タイプ | isIdenticalTo | |||||||||||
識別子タイプ | NAID | |||||||||||
関連識別子 | 110006991259 | |||||||||||
ISSN | ||||||||||||
収録物識別子タイプ | PISSN | |||||||||||
収録物識別子 | 1345-1529 | |||||||||||
著作権関連情報 | ||||||||||||
権利情報 | Copyright©1995 Biomedical Fuzzy Systems Association. 本文データは学協会の許諾に基づきCiNiiから複製したものである | |||||||||||
キーワード | ||||||||||||
主題Scheme | Other | |||||||||||
主題 | Associative memory | |||||||||||
キーワード | ||||||||||||
主題Scheme | Other | |||||||||||
主題 | bottleneck neural network | |||||||||||
キーワード | ||||||||||||
主題Scheme | Other | |||||||||||
主題 | multilayer perceptron | |||||||||||
キーワード | ||||||||||||
主題Scheme | Other | |||||||||||
主題 | intermediate element | |||||||||||
キーワード | ||||||||||||
主題Scheme | Other | |||||||||||
主題 | noise tolerance | |||||||||||
出版タイプ | ||||||||||||
出版タイプ | VoR | |||||||||||
出版タイプResource | http://purl.org/coar/version/c_970fb48d4fbd8a85 | |||||||||||
査読の有無 | ||||||||||||
値 | yes | |||||||||||
業績ID | ||||||||||||
6E92DB0348374B3B4925762B001D34B5 | ||||||||||||
著者別名 | ||||||||||||
姓名 | 横井, 博一 | |||||||||||
言語 | ja |