關於此課程：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 second course of the two would focus more on algorithmic tools, and the other course would focus more on mathematical 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。本課程將較為著重方法類的工具，而另一課程將較為著重數學類的工具。]
製作方： National Taiwan University
教學方： 林軒田, 教授 (Professor)
資訊工程學系 (Computer Science and Information Engineering)
| 級別 | 中級 |
| 語言 | Chinese (Traditional) |
| 硬件要求 | 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 |
| 如何通過 | 通過所有計分作業以完成課程。 |
| 用戶評分 |
授課大綱
第 1 週
第九講: Linear Regression
weight vector for linear hypotheses and squared error instantly calculated by analytic solution
4 個視頻, 4 個閱讀材料
- Материал для самостоятельного изучения: NTU MOOC 課程問題詢問與回報機制
- 視頻: Linear Regression Algorithm
- 視頻: Generalization Issue
- 視頻: Linear Regression for Binary Classification
- Материал для самостоятельного изучения: 課程大綱
- Материал для самостоятельного изучения: 課程形式及評分標準
- Материал для самостоятельного изучения: 延伸閱讀
第 2 週
第十講: Logistic Regression
gradient descent on cross-entropy error
to get good logistic hypothesis
4 個視頻
- 視頻: Logistic Regression Error
- 視頻: Gradient of Logistic Regression Error
- 視頻: Gradient Descent
第 3 週
第十一講: Linear Models for Classification
binary classification via (logistic) regression; multiclass classification via OVA/OVO decomposition
4 個視頻
- 視頻: Stochastic Gradient Descent
- 視頻: Multiclass via Logistic Regression
- 視頻: Multiclass via Binary Classification
第 4 週
第十二講: Nonlinear Transformation
nonlinear model via nonlinear feature transform+linear model with price of model complexity
4 個視頻
- 視頻: Nonlinear Transform
- 視頻: Price of Nonlinear Transform
- 視頻: Structured Hypothesis Sets
已評分: 作業三
第 5 週
第十三講: Hazard of Overfitting
overfitting happens with excessive power, stochastic/deterministic noise and limited data
4 個視頻
- 視頻: The Role of Noise and Data Size
- 視頻: Deterministic Noise
- 視頻: Dealing with Overfitting
第 6 週
第十四講: Regularization
minimize augmented error, where the added regularizer effectively limits model complexity
4 個視頻
- 視頻: Weight Decay Regularization
- 視頻: Regularization and VC Theory
- 視頻: General Regularizers
第 7 週
第十五講: Validation
(crossly) reserve validation data to simulate testing procedure for model selection
4 個視頻
- 視頻: Validation
- 視頻: Leave-One-Out Cross Validation
- 視頻: V-Fold Cross Validation
第 8 週
第十六講: Three Learning Principles
be aware of model complexity, data goodness and your professionalism
4 個視頻
- 視頻: Sampling Bias
- 視頻: Data Snooping
- 視頻: Power of Three
已評分: 作業四
常見問題解答
運作方式
Задания курса
Каждый курс — это интерактивный учебник, который содержит видеоматериалы, тесты и проекты.
Помощь сокурсников
Общайтесь с тысячами других учащихся: обсуждайте идеи, материалы курса и помогайте друг другу осваивать новые понятия.
Сертификаты
Получите документы о прохождении курсов и поделитесь своим успехом с друзьями, коллегами и работодателями.
製作方
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.More courses information, the official Facebook Page: https://www.facebook.com/ntumooc2017/
評分和審閱
内容相对上部要容易，涉及到的算法比较基本。
Best ML course I have ever taken!
谢谢老师助教们的付出！
蕭
A great course