This course presents a unified treatment of machine learning problems and

solutions. Basically, machine learning is about programming computers to

optimize a performance criterion using example data or past experience. Consider

the recognition of spoken speech—that is, converting the acoustic speech signal

to an ASCII text; humans can do this task seemingly without any difficulty, but

we are unable to explain how we do it. In machine learning, the approach is to

collect a large collection of sample utterances from different people and learn

to map these to words. Another example is that developers of a web site (e.g.,

YouTube) usually collect user behaviors (e.g., mouse clicks), apply machine

learning to analyze the preference of individual users, and recommend items

(e.g., clips) that may be interesting to these users.

The topics to be discussed have their bases in different fields, including

statistics, pattern recognition, neural networks, artificial intelligence,

signal processing, control, and data mining. All algorithms are explained so

that the student can easily move from the equations to a computer program.

This course is intended for senior undergraduate and graduate students who have

proper understanding of computer programming, probability, calculus, and linear

algebra.

二、 指定用書 (Textbook)

[1]. Lecture Notes

[2]. Introduction to Machine Learning, 2ed,by Ethem Alpaydin, ISBN:

026201243X

三、 參考書籍 (References)

[1]. Pattern Recognition and Machine Learning, by Christopher M. Bishop,

ISBN: 0387310738

[2]. Data Mining: Practical Machine Learning Tools and Techniques, 2ed, by

Ian H. Witten, et al., ISBN: 0120884070

四、 教學方式 (Teaching Method)

Lecture

五、 教學進度 (Syllabus)

1. Introduction to machine learning

2. A review of the basics: linear algebra, multivariate calculus, and

probability

3. Supervise learning

4. Bayesian decision theory

5. Parametric methods

6. Multivariate methods

7. Dimensionality reduction

8. Clustering

9. Nonparametric methods

10. Linear discrimination

11. Kernel machines

12. Hidden Markov models

13. Graphical models

14. Combining multiple leaners

15. Metrics and experiments

六、 成績考核 (Evaluation)

Midterm exam: 30%

Final exam: 40%

Assignments 30%