About this course: Machine learning is the study that allows computers to adaptively improve their performance with experience accumulated from the data observed. Our two sister courses teach the most fundamental algorithmic, theoretical and practical tools that any user of machine learning needs to know. This first course of the two would focus more on mathematical tools, and the other course would focus more on algorithmic tools. [機器學習旨在讓電腦能由資料中累積的經驗來自我進步。我們的兩項姊妹課程將介紹各領域中的機器學習使用者都應該知道的基礎演算法、理論及實務工具。本課程將較為著重數學類的工具，而另一課程將較為著重方法類的工具。]
Created by: National Taiwan University
Taught by: Hsuan-Tien Lin, 林軒田, 副教授 (Associate Professor)
資訊工程學系 (Computer Science and Information Engineering)
| Level | Beginner |
| Language | Chinese (Traditional) |
| Hardware Req | 需要一般個人電腦或同等級的計算資源以撰寫程式執行作業 |
| How To Pass | Pass all graded assignments to complete the course. |
| User Ratings |
Syllabus
WEEK 1
The Learning Problem
what machine learning is and its connection to applications and other fields
5 videos, 4 readings
- Video: Course Introduction
- Video: What is Machine Learning
- Video: Applications of Machine Learning
- Video: Components of Machine Learning
- Video: Machine Learning and Other Fields
- Reading: 課程大綱
- Reading: 課程形式及評分標準
- Reading: 延伸閱讀
- Reading: homework 0
WEEK 2
第二講：Learning to Answer Yes/No
your first learning algorithm (and the world's first!) that "draws the line" between yes and no by adaptively searching for a good line based on data
4 videos
- Video: Perceptron Hypothesis Set
- Video: Perceptron Learning Algorithm (PLA)
- Video: Guarantee of PLA
- Video: Non-Separable Data
WEEK 3
第三講：Types of Learning
learning comes with many possibilities in different applications, with our focus being binary classification or regression from a batch of supervised data with concrete features
4 videos
- Video: Learning with Different Output Space
- Video: Learning with Different Data Label
- Video: Learning with Different Protocol
- Video: Learning with Different Input Space
WEEK 4
第四講：Feasibility of Learning
learning can be "probably approximately correct" when given enough statistical data and finite number of hypotheses
4 videos
- Video: Learning is Impossible?
- Video: Probability to the Rescue
- Video: Connection to Learning
- Video: Connection to Real Learning
Graded: 作業一
WEEK 5
第五講：Training versus Testing
what we pay in choosing hypotheses during training: the growth function for representing effective number of choices
4 videos
- Video: Recap and Preview
- Video: Effective Number of Lines
- Video: Effective Number of Hypotheses
- Video: Break Point
WEEK 6
第六講: Theory of Generalization
test error can approximate training error if there is enough data and growth function does not grow too fast
4 videos
- Video: Restriction of Break Point
- Video: Bounding Function: Basic Cases
- Video: Bounding Function: Inductive Cases
- Video: A Pictorial Proof
WEEK 7
第七講: The VC Dimension
learning happens if there is finite model complexity (called VC dimension), enough data, and low training error
4 videos
- Video: Definition of VC Dimension
- Video: VC Dimension of Perceptrons
- Video: Physical Intuition of VC Dimension
- Video: Interpreting VC Dimension
WEEK 8
第八講: Noise and Error
learning can still happen within a noisy environment and different error measures
4 videos
- Video: Noise and Probabilistic Target
- Video: Error Measure
- Video: Algorithmic Error Measure
- Video: Weighted Classification
Graded: 作業二
FAQs
How It Works
Coursework
Each course is like an interactive textbook, featuring pre-recorded videos, quizzes and projects.
Help from Your Peers
Connect with thousands of other learners and debate ideas, discuss course material, and get help mastering concepts.
Certificates
Earn official recognition for your work, and share your success with friends, colleagues, and employers.
Creators
National Taiwan University
We firmly believe that open access to learning is a powerful socioeconomic equalizer. NTU is especially delighted to join other world-class universities on Coursera and to offer quality university courses to the Chinese-speaking population. We hope to transform the rich rewards of learning from a limited commodity to an experience available to all.
Ratings and Reviews
許多名詞似乎是自創新詞，但都能很好地描述ML的理論
課程的統計很吃重，難度的分配有些不均勻
整體來說是非常適合有數學底子學生的扎實入門課程
课程逻辑严密
林老师讲课很好，但是希望老师在讲到一些比较容易混淆的概念的时候，可以举一些例子帮助我们加深理解。
Coursera provides universal access to the world’s best education, partnering with top universities and organizations to offer courses online.
© 2017 Coursera Inc. All rights reserved.
作业题好难，讲的课毕竟还是最简单的概念。