政大機構典藏-National Chengchi University Institutional Repository(NCCUR):Item 140.119/156614
English  |  正體中文  |  简体中文  |  Post-Print筆數 : 27 |  全文筆數/總筆數 : 115418/146445 (79%)
造訪人次 : 55127895      線上人數 : 16
RC Version 6.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
搜尋範圍 查詢小技巧:
  • 您可在西文檢索詞彙前後加上"雙引號",以獲取較精準的檢索結果
  • 若欲以作者姓名搜尋,建議至進階搜尋限定作者欄位,可獲得較完整資料
  • 進階搜尋
    請使用永久網址來引用或連結此文件: https://nccur.lib.nccu.edu.tw/handle/140.119/156614


    題名: 利用深度學習之流形重建之研究
    Manifold Reconstruction with Deep Learning
    作者: 蔡炎龍
    貢獻者: 應數系
    關鍵詞: 深度學習;流形重建;惠特尼擴問題;黎曼流形
    Deep Learning;Manifold Reconstruction;Whitney’s extension problem;Riemannian manifolds
    日期: 2021-10
    上傳時間: 2025-04-16 14:28:19 (UTC+8)
    摘要: 本計畫為「新興或跨領域計畫」。我們以 Fefferman 最近流形重建的文章為藍本, 試圖以數學的角度去探討深度學習的問題。本計畫以深度學習, 並用一個類似 ResNet 的結構, 去取代 Fefferman 文章中演算法用到的重要部份。希望同樣在給定的條件之下, 我們也能用類似的手法, 透過深度學習建構一個合乎要求的流形。更進一步的, 我們希望能用更數學的語言, 去描述深度學習。其中一個例子我們希望運用「蒸餾」的手法, 討論是否可能找到某種意涵下的「最小模型」。
    We are based on Fefferman's recent article on manifold reconstruction as an attempt to explore the theories of deep learning from a mathematical perspective. This project uses deep learning and a ResNet-like structure to replace the ones used in the algorithm in Fefferman's article. It is hoped that under the given conditions, we can also use a similar method to construct a desirable manifold through deep learning. Furthermore, we hope to describe deep learning in more mathematics way. For instance, we want to use the method of "distillation" to discuss whether it is possible to find a "minimum model" in some sense.
    關聯: 科技部, MOST109-2115-M004-011, 109.08-110.07
    資料類型: report
    顯示於類別:[應用數學系] 國科會研究計畫

    文件中的檔案:

    檔案 描述 大小格式瀏覽次數
    index.html0KbHTML9檢視/開啟


    在政大典藏中所有的資料項目都受到原著作權保護.


    社群 sharing

    著作權政策宣告 Copyright Announcement
    1.本網站之數位內容為國立政治大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,惟仍請適度,合理使用本網站之內容,以尊重著作權人之權益。商業上之利用,則請先取得著作權人之授權。
    The digital content of this website is part of National Chengchi University Institutional Repository. It provides free access to academic research and public education for non-commercial use. Please utilize it in a proper and reasonable manner and respect the rights of copyright owners. For commercial use, please obtain authorization from the copyright owner in advance.

    2.本網站之製作,已盡力防止侵害著作權人之權益,如仍發現本網站之數位內容有侵害著作權人權益情事者,請權利人通知本網站維護人員(nccur@nccu.edu.tw),維護人員將立即採取移除該數位著作等補救措施。
    NCCU Institutional Repository is made to protect the interests of copyright owners. If you believe that any material on the website infringes copyright, please contact our staff(nccur@nccu.edu.tw). We will remove the work from the repository and investigate your claim.
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - 回饋