NTU Course
NewsHelpOverview

Machine Learning in Atmospheric Thermodynamics

Offered in 111-1
  • Serial Number

    10591

  • Course Number

    AtmSci8049

  • Course Identifier

    229 D3130

  • No Class

  • 3 Credits
  • Elective

    GRADUATE INSTITUTE OF ATMOSPHERIC SCIENCES

      Elective
    • GRADUATE INSTITUTE OF ATMOSPHERIC SCIENCES

  • WU CHIEN-MING
  • Fri 2, 3, 4
  • 大氣A104

  • Type 3

  • 10 Student Quota

    NTU 10

  • No Specialization Program

  • Chinese
  • Core Capabilities and Curriculum Planning
  • Notes
    Not open in course pre-registration period。
  • NTU Enrollment Status

    Enrolled
    0/10
    Other Depts
    0/0
    Remaining
    0
    Registered
    0
  • Course Description
    本課程主旨在討論大氣熱力學原理如何應用於機器學習之資料分析方法,包括如何以符合物理原則之方式進行資料前處理,以及透過物理詮釋的觀點評估機器學習模型之表現。課程內容將會使用再分析資料(ERA5 reanalysis)及高解析雲解析模式(cloud-resolving model, CRM)之模擬結果,針對特定之大氣現象進行機器學習模型建構。 This course aims to discuss how to apply machine learning techniques to analyze atmospheric data from the perspective of atmospheric thermodynamics, which includes the physical-based preprocessing of data and the physical evaluation of the performance of the neural network models. The ERA5 reanalysis dataset and the cloud-resolving model outputs will be taken as the input data to apply the physical-based machine learning techniques to understand the specific atmospheric phenomena.
  • Course Objective
    A. 理解大氣熱力學中的關鍵議題 B. 大氣資料的處理及視覺化 C. 機器學習的基本原理及實作應用 D. 對機器學習結果的物理詮釋
  • Course Requirement
    A. 具有大氣熱力學基礎知識,並有資料分析與視覺化之程式能力。建議學生能先修大學部之流體力學、熱力學、動力學、計算機語言、數值分析等課程 B. 須能依據作業要求進行大氣數據分析及機器學習模型程式撰寫,並參與課堂報告及討論。
  • Expected weekly study hours after class
  • Office Hour
    *This office hour requires an appointment
  • Designated Reading
  • References
  • Grading
  • Adjustment methods for students
  • Course Schedule
    Week 1
    Week 2
    Week 3
    Week 4
    Week 5
    Week 6
    Week 7
    Week 8
    Week 9
    Week 10
    Week 11
    Week 12
    Week 13
    Week 14
    Week 15
    Week 16
    Week 17