WEKO3
アイテム
{"_buckets": {"deposit": "a694ee0f-e301-4fdd-8ff8-e25a96a6ab37"}, "_deposit": {"id": "1852", "owners": [], "pid": {"revision_id": 0, "type": "depid", "value": "1852"}, "status": "published"}, "_oai": {"id": "oai:niigata-u.repo.nii.ac.jp:00001852", "sets": ["454", "425"]}, "item_5_biblio_info_6": {"attribute_name": "書誌情報", "attribute_value_mlt": [{"bibliographicIssueDates": {"bibliographicIssueDate": "2013-05", "bibliographicIssueDateType": "Issued"}, "bibliographicIssueNumber": "5", "bibliographicPageEnd": "939", "bibliographicPageStart": "935", "bibliographicVolumeNumber": "E96-A", "bibliographic_titles": [{"bibliographic_title": "IEICE transactions on fundamentals of electronics, communications and computer sciences"}, {"bibliographic_title": "IEICE transactions on fundamentals of electronics, communications and computer sciences", "bibliographic_titleLang": "en"}]}]}, "item_5_description_4": {"attribute_name": "抄録", "attribute_value_mlt": [{"subitem_description": "This work proposes an exponential computation with low-computational complexity and applies this technique to the expectation-maximization (EM) algorithm for Gaussian mixture model (GMM). For certain machine-learning techniques, such as the EM algorithm for the GMM, fast and low-cost implementations are preferred over high precision ones. Since the exponential function is frequently used in machine-learning algorithms, this work proposes reducing computational complexity by transforming the function into powers of two and introducing a look-up table. Moreover, to improve efficiency the look-up table is scaled. To verify the validity of the proposed technique, this work obtains simulation results for the EM algorithm used for parameter estimation and evaluates the performances of the results in terms of the mean absolute error and computational time. This work compares our proposed method against the Taylor expansion and the exp( ) function in a standard C library, and shows that the computational time of the EM algorithm is reduced while maintaining comparable precision in the estimation results.", "subitem_description_type": "Abstract"}]}, "item_5_publisher_7": {"attribute_name": "出版者", "attribute_value_mlt": [{"subitem_publisher": "The Institute of Electronics, Information and Communication Engineers"}]}, "item_5_relation_14": {"attribute_name": "DOI", "attribute_value_mlt": [{"subitem_relation_type_id": {"subitem_relation_type_id_text": "info:doi/10.1587/transfun.E96.A.935", "subitem_relation_type_select": "DOI"}}]}, "item_5_relation_31": {"attribute_name": "異版である", "attribute_value_mlt": [{"subitem_relation_type": "isVersionOf", "subitem_relation_type_id": {"subitem_relation_type_id_text": "http://www.ieice.org/jpn/trans_online/", "subitem_relation_type_select": "URI"}}]}, "item_5_rights_15": {"attribute_name": "権利", "attribute_value_mlt": [{"subitem_rights": "Copyright(C)2013 IEICE"}]}, "item_5_select_19": {"attribute_name": "著者版フラグ", "attribute_value_mlt": [{"subitem_select_item": "publisher"}]}, "item_5_source_id_11": {"attribute_name": "書誌レコードID", "attribute_value_mlt": [{"subitem_source_identifier": "AA10826239", "subitem_source_identifier_type": "NCID"}]}, "item_5_source_id_9": {"attribute_name": "ISSN", "attribute_value_mlt": [{"subitem_source_identifier": "09168508", "subitem_source_identifier_type": "ISSN"}]}, "item_creator": {"attribute_name": "著者", "attribute_type": "creator", "attribute_value_mlt": [{"creatorNames": [{"creatorName": "Watanabe, Hidenori"}], "nameIdentifiers": [{"nameIdentifier": "5863", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Muramatsu, Shogo"}], "nameIdentifiers": [{"nameIdentifier": "5864", "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": "2019-07-29"}], "displaytype": "detail", "download_preview_message": "", "file_order": 0, "filename": "E96-A_5_935-939.pdf", "filesize": [{"value": "715.2 kB"}], "format": "application/pdf", "future_date_message": "", "is_thumbnail": false, "licensetype": "license_free", "mimetype": "application/pdf", "size": 715200.0, "url": {"label": "E96-A_5_935-939.pdf", "url": "https://niigata-u.repo.nii.ac.jp/record/1852/files/E96-A_5_935-939.pdf"}, "version_id": "d7d861d6-540a-4099-a805-97bb2745e91f"}]}, "item_keyword": {"attribute_name": "キーワード", "attribute_value_mlt": [{"subitem_subject": "EM algorithm", "subitem_subject_scheme": "Other"}, {"subitem_subject": "Gaussian mixture model", "subitem_subject_scheme": "Other"}, {"subitem_subject": "exponential function", "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": "Look-Up-Table-Based Exponential Computation and Application to an EM Algorithm for GMM", "item_titles": {"attribute_name": "タイトル", "attribute_value_mlt": [{"subitem_title": "Look-Up-Table-Based Exponential Computation and Application to an EM Algorithm for GMM"}, {"subitem_title": "Look-Up-Table-Based Exponential Computation and Application to an EM Algorithm for GMM", "subitem_title_language": "en"}]}, "item_type_id": "5", "owner": "1", "path": ["454", "425"], "permalink_uri": "http://hdl.handle.net/10191/30659", "pubdate": {"attribute_name": "公開日", "attribute_value": "2014-12-02"}, "publish_date": "2014-12-02", "publish_status": "0", "recid": "1852", "relation": {}, "relation_version_is_last": true, "title": ["Look-Up-Table-Based Exponential Computation and Application to an EM Algorithm for GMM"], "weko_shared_id": null}
Look-Up-Table-Based Exponential Computation and Application to an EM Algorithm for GMM
http://hdl.handle.net/10191/30659
http://hdl.handle.net/10191/3065990b23a86-66b7-49b6-865d-af5b1093c431
名前 / ファイル | ライセンス | アクション |
---|---|---|
![]() |
|
Item type | 学術雑誌論文 / Journal Article(1) | |||||
---|---|---|---|---|---|---|
公開日 | 2014-12-02 | |||||
タイトル | ||||||
タイトル | Look-Up-Table-Based Exponential Computation and Application to an EM Algorithm for GMM | |||||
タイトル | ||||||
言語 | en | |||||
タイトル | Look-Up-Table-Based Exponential Computation and Application to an EM Algorithm for GMM | |||||
言語 | ||||||
言語 | eng | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | EM algorithm | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | Gaussian mixture model | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | exponential function | |||||
資源タイプ | ||||||
資源 | http://purl.org/coar/resource_type/c_6501 | |||||
タイプ | journal article | |||||
著者 |
Watanabe, Hidenori
× Watanabe, Hidenori× Muramatsu, Shogo |
|||||
抄録 | ||||||
内容記述タイプ | Abstract | |||||
内容記述 | This work proposes an exponential computation with low-computational complexity and applies this technique to the expectation-maximization (EM) algorithm for Gaussian mixture model (GMM). For certain machine-learning techniques, such as the EM algorithm for the GMM, fast and low-cost implementations are preferred over high precision ones. Since the exponential function is frequently used in machine-learning algorithms, this work proposes reducing computational complexity by transforming the function into powers of two and introducing a look-up table. Moreover, to improve efficiency the look-up table is scaled. To verify the validity of the proposed technique, this work obtains simulation results for the EM algorithm used for parameter estimation and evaluates the performances of the results in terms of the mean absolute error and computational time. This work compares our proposed method against the Taylor expansion and the exp( ) function in a standard C library, and shows that the computational time of the EM algorithm is reduced while maintaining comparable precision in the estimation results. | |||||
書誌情報 |
IEICE transactions on fundamentals of electronics, communications and computer sciences en : IEICE transactions on fundamentals of electronics, communications and computer sciences 巻 E96-A, 号 5, p. 935-939, 発行日 2013-05 |
|||||
出版者 | ||||||
出版者 | The Institute of Electronics, Information and Communication Engineers | |||||
ISSN | ||||||
収録物識別子タイプ | ISSN | |||||
収録物識別子 | 09168508 | |||||
書誌レコードID | ||||||
収録物識別子タイプ | NCID | |||||
収録物識別子 | AA10826239 | |||||
DOI | ||||||
識別子タイプ | DOI | |||||
関連識別子 | info:doi/10.1587/transfun.E96.A.935 | |||||
権利 | ||||||
権利情報 | Copyright(C)2013 IEICE | |||||
著者版フラグ | ||||||
値 | publisher | |||||
異版である | ||||||
関連タイプ | isVersionOf | |||||
識別子タイプ | URI | |||||
関連識別子 | http://www.ieice.org/jpn/trans_online/ |