投稿

林軒田(Hsuan-Tien Lin) - [機器學習基石]Machine Learning Foundations -

9.9万

2627

2014-10-16 00:13:24

課程對應到台灣大學所開設之「機器學習」課程的前半學期，課程的強度比照世界一流大學的實體課程設計，希望能幫大家打下紥實的基礎，所以在份量上會比常見的線上課程要吃重一些，建議大家要預留充足的時間來把這門課學好。即使有前面這句話的「警告」，這門課在台大依然有著很好的口碑，因此這幾年一直看到同學們「前仆後繼」地來修習，希望能和台大的同學一樣喜愛這門課程。讲师:林軒田(Hsuan-Tien Lin)

MACHINE LEARNING

Dauglas 发消息

人之所以强大,是因为领悟了悲伤.-独立游戏《泠：落日孤行》制作者-在职技术策划-资浅海鸥球迷

1 - 1 - Course Introduction

10:58

1 - 2 - What is Machine Learning

18:29

1 - 3 - Applications of Machine Learning

18:57

1 - 4 - Components of Machine Learning

11:45

1 - 5 - Machine Learning and Other Fields

10:21

2 - 1 - Perceptron Hypothesis Set

15:42

2 - 2 - Perceptron Learning Algorithm (PLA)

19:46

2 - 3 - Guarantee of PLA

12:37

2 - 4 - Non-Separable Data

12:55

3 - 1 - Learning with Different Output Space

17:26

3 - 2 - Learning with Different Data Label

18:12

3 - 3 - Learning with Different Protocol

11:09

3 - 4 - Learning with Different Input Space

14:13

4 - 1 - Learning is Impossible

13:32

4 - 2 - Probability to the Rescue

11:32

4 - 3 - Connection to Learning

16:46

4 - 4 - Connection to Real Learning

18:05

5 - 1 - Recap and Preview

13:44

5 - 2 - Effective Number of Lines

15:26

5 - 3 - Effective Number of Hypotheses

16:17

5 - 4 - Break Point

07:44

6 - 1 - Restriction of Break Point

14:18

6 - 2 - Bounding Function Basic Cases

06:56

6 - 3 - Bounding Function Inductive Cases

14:46

6 - 4 - A Pictorial Proof

16:01

7 - 1 - Definition of VC Dimension

13:10

7 - 2 - VC Dimension of Perceptrons

13:27

7 - 3 - Physical Intuition of VC Dimension

06:10

7 - 4 - Interpreting VC Dimension

17:13

8 - 1 - Noise and Probabilistic Target

17:01

8 - 2 - Error Measure

15:10

8 - 3 - Algorithmic Error Measure

13:47

8 - 4 - Weighted Classification (16-54)

16:55

9 - 1 - Linear Regression Problem

10:08

9 - 2 - Linear Regression Algorithm (20-03)

20:04

9 - 3 - Generalization Issue

20:34

9 - 4 - Linear Regression for Binary Classification

11:23

10 - 1 - Logistic Regression Problem

14:32

10 - 2 - Logistic Regression Error

15:58

10 - 3 - Gradient of Logistic Regression Error

15:37

10 - 4 - Gradient Descent

19:18

11 - 1 - Linear Models for Binary Classification

21:35

11 - 2 - Stochastic Gradient Descent

11:39

11 - 3 - Multiclass via Logistic Regression

14:18

11 - 4 - Multiclass via Binary Classification

11:35

12 - 1 - Quadratic Hypothesis (23-47)

23:48

12 - 2 - Nonlinear Transform (09-52)

09:53

12 - 3 - Price of Nonlinear Transform (15-37)

15:38

12 - 4 - Structured Hypothesis Sets (09-36)

09:36

13 - 1 - What is Overfitting- (10-45)

10:45

13 - 2 - The Role of Noise and Data Size (13-36)

13:36

13 - 3 - Deterministic Noise (14-07)

14:08

13 - 4 - Dealing with Overfitting (10-49)

10:50

14 - 1 - Regularized Hypothesis Set (19-16)

19:17

14 - 2 - Weight Decay Regularization (24-08)

24:09

14 - 3 - Regularization and VC Theory (08-15)

08:16

14 - 4 - General Regularizers (13-28)

13:29

15 - 1 - Model Selection Problem (16-00)

16:01

15 - 2 - Validation (13-24)

13:25

15 - 3 - Leave-One-Out Cross Validation (16-06)

16:07

15 - 4 - V-Fold Cross Validation (10-41)

10:42

16 - 1 - Occam-'s Razor (10-08)

10:09

16 - 2 - Sampling Bias (11-50)

11:51

16 - 3 - Data Snooping (12-28)

12:29

16 - 4 - Power of Three (08-49)

08:50

顶部