A Robust Self-Constructing Normalized Gaussian Network for Online Machine Learning

• Material Type: 博士論文
• Title: A Robust Self-Constructing Normalized Gaussian Network for Online Machine Learning
• Author/Editor: Backhus, Jana Cathrin
• Author Heading: Backhus, Jana Cathrin
• Publication, Distribution, etc.: Hokkaido University
• Publication Date: 2017-03-23
• Publication Date (W3CDTF): 2017-03-23
• Alternative Title: オンライン機械学習のための頑健な自己構築正規化ガウスネットワーク
• Contributor: 杉本, 雅則; 工藤, 峰一; 今井, 英幸; 瀧川, 一学
• Degree grantor/type: 北海道大学
• Date Granted: 2017-03-23
• Date Granted (W3CDTF): 2017-03-23
• Dissertation Number: 甲第12624号
• Degree Type: 博士(情報科学)
• Conferring No. (Dissertation): 甲第12624号
• Text Language Code: eng
• NDC: 500
• Target Audience: 一般
• Note (General): In this thesis, I aim to improve the robustness and applicability of Normalized Gaussian networks (NGnet) in the context of online machine learning tasks. A challenging problem in online machine learning is the limited domain knowledge provoked by restricted prior knowledge, while additional information are received only sequentially over time. The limited domain knowledge makes the application of artificial neural networks (ANN) more difficult, and major challenges include negative interference and the selection of an accurate model complexity. In this thesis, I consider these challenges in regard to the NGnet, which belongs to a group of ANNs that possess a certain grade of robustness against negative interference due to the local properties of their network architecture. Yet, further improvements of robustness are necessary in regard to the ANN’s learning algorithm and model complexity selection. A recently proposed learning algorithm with localized forgetting provides robustness against negative interference, but it is not applicable over the full numerical range of an implied discount factor. Also, dynamic model selection was yet to be considered. Therefore, I revise the localized forgetting approach and adapt dynamic model selection to it in a self-constructing manner. Dynamic model selection has been considered for an earlier learning algorithm of the NGnet with global forgetting, which however shows a non-robust behavior in negative interference prone environments. Then, I propose localization of some of the model selection mechanisms to improve their robustness and add a new merge manipulation to deal with model redundancies. The effectiveness of the proposed method is compared with earlier learning approaches of the NGnet for several experiments. The proposed method possesses robust and favorable performance in the different tested learning environments, making it the better alternative when applied to online learning tasks with proneness to negative interference.; (主査) 教授 杉本 雅則, 教授 工藤 峰一, 教授 今井 英幸, 准教授 瀧川 一学; 情報科学研究科（情報理工学専攻）
• DOI: 10.14943/doctoral.k12624; https://doi.org/10.14943/doctoral.k12624
• Persistent ID (NDL): info:ndljp/pid/10370873; https://dl.ndl.go.jp/pid/10370873
• Collection: 障害者向け資料
• Collection (Materials For Handicapped People:1): テキストデータ
• Collection (particular): 国立国会図書館デジタルコレクション > デジタル化資料 > 博士論文; https://dl.ndl.go.jp/collections/A00014
• Acquisition Basis: 博士論文（自動収集）
• Date Accepted (W3CDTF): 2017-07-03T04:10:06+09:00
• Date Created (W3CDTF): 2017-02-14
• Format (IMT): PDF
• Access Restrictions: 国立国会図書館内限定公開
• Service for the Digitized Contents Transmission Service: 図書館・個人送信対象外
• Availability of remote photoduplication service: 可
• Periodical Title (URI): http://dx.doi.org/10.14943/doctoral.k12624; http://hdl.handle.net/2115/65449
• Is Referenced By (URI): http://hdl.handle.net/2115/65447
• Data Provider (Database): 国立国会図書館 : 国立国会図書館デジタルコレクション; https://dl.ndl.go.jp