課程資訊
課程名稱
機器學習中的數學原理
Mathematical Principles of Machine Learning 
開課學期
106-2 
授課對象
電機資訊學院  電機工程學研究所  
授課教師
王奕翔 
課號
CommE5051 
課程識別碼
942 U0650 
班次
 
學分
3.0 
全/半年
半年 
必/選修
選修 
上課時間
星期三7,8,9(14:20~17:20) 
上課地點
明達231 
備註
總人數上限:60人 
課程網頁
http://homepage.ntu.edu.tw/~ihwang/Teaching/Sp18/MPML.html 
課程簡介影片
 
核心能力關聯
核心能力與課程規劃關聯圖
課程大綱
為確保您我的權利,請尊重智慧財產權及不得非法影印
課程概述

This course aims to introduce the mathematical principles of machine learning. The course is roughly divided into two parts: the statistical principles, and the optimization principles. For the former, we will focus on introducing basics of statistical learning theory, where the main focus is what can be learned and how good can be learned. For the latter, we will focus on algorithmic aspects of optimization, which play a key role in machine learning. As for theoretical topics, tentatively we aim to cover VC dimension, Rademacher complexity, and oracle complexity. As for learning models, we aim to cover SVM, neural networks, and boosting. 

課程目標
1. Introduce main concepts underlying machine learning with mathematical rigor.
2. Uncover mathematical principles underlying various machine learning techniques.
3. Introduce methods to theoretically analyze learning algorithms.
4. Develop theory-oriented thinking which helps understand existing algorithms and create novel ones. 
課程要求
Prerequisite: Calculus, Probability, Linear Algebra.
Preferable (but optional): Machine learning, Convex optimization, Real analysis.
Grading: Exam (20%), Homework (40%), Project (40%) 
預期每週課後學習時數
 
Office Hours
 
參考書目
1. Shai Shalev-Shwartz and Shai Ben-David, Understanding Machine Learning: From Theory to Algorithms, Cambridge University Press, 2014.
2. Mehryar Mohri, Afshin Rostamizadeh, and Ameet Talwalkar, Foundations of Machine Learning, the MIT Press, 2012.
3. S. Boyd and L. Vandenberghe, Convex Optimization, Cambridge University Press, 2004.
4. Y. Nesterov, Introductory lectures on convex optimization: A basic course. Kluwer Academic Publishers, 2004.
5. Additional references: research papers and surveys to be assigned during lectures. 
指定閱讀
Lectures will be based on lecture notes and slides. 
評量方式
(僅供參考)
 
No.
項目
百分比
說明
1. 
no 
100% 
 
 
課程進度
週次
日期
單元主題