Office hours: Wed. 4:10-5:30pm (N210, Zibin building)
Instructor: Yanwei Fu (yanweifu@fudan.edu.cn)
Teaching Assistants: (各位同学,有什么事情,可以给两个助教发邮件,也可以wechat联系助教.
Please email/Wechat the two TAs if you have any problems.)
(1)谢宇 wechat: Y1314941 email: 15955038579@163.com
(2)孙强 wechat: sunqiang6861 email: sunqiang85@gmail.com
Synopsis: As an introduction to statistical learning and machine learning, this course is about learning from data: statistical learning refers to a set of tools for modeling and understanding complex datasets; and machine learning is defined as a set of methods that can automatically detect patterns in data, and then use the uncovered patterns to predict future data, or to perform other kinds of decision making under uncertainty. Thus, the main objectiveness is to present students a unified view of both two fields through the teaching of the methodology, applications and the key ideas behind the methods. The whole course is illustrated with R as well as other statistical programming languages such as Matlab and Python. We aim at gradually cultivating students the abilities of both theoretical analysis and practical problem solving.
Textbook:
1. James, Witten, Hastie and Tibshirani (2013) An
Introduction to Statistical Learning, with applications in R. Springer.
2. Bishop, C.M. (2006), Pattern recognition and Machine Learning, Springer.
3. Hastie, T., Tibshurani, R. and Friedman, J. (2011)
The Elements of Statistical Learning, data mining, inference and
Prediction, 2ndEdition. Springer.
Prerequisites:
Registration:
Grading:
(1) Class attendance (10%), includes class performance, class discussion and critical thinking.
Each
absent: -1%
(2) Weekly homework (20%), is of 5 times. We expect the student can finish each one within 1.5-2.5 hours.
Each:
4% *5;
Late
Submission (after Dec. 14th, 2017) will be penalized by 50% of total
score of each homework, in other word, the highest score for each one
of the late submission is 2%.
(3) Monthly mini-projects (50%), is of 3-4 projects which are selected from the real-world Big-Data problems, including but not limited to, computer vision, pattern recognition, recommendation system, social network, financial data analysis and bioinformatics. In general, the reports should be written in English, and include algorithm skims (3%), critical codes (2%), experimental analysis (3%); and the discussion of proposed method (2%).
About the Submission of mini-projects.
The
report can be written by Word or Latex. Generate a single pdf file
of your mini-projects. The file name should be
SLML_yourname_student-id.pdf. Also put the names and Student ID in
your paper. To submit the report, email the pdf file to 15955038579@163.com.
About the deadline and penalty. In general, you should submit the
paper according to the deadline of each mini-project. The late submission
is also acceptable; however, you will be penalized 10% of scores for each
week's delay.
(4)
Final project (20%) is finished by one team. Each team should have up to
3 students; and will solve a real-world Big-Data problem. In
general, the final report should be written in English. The
main components of the report will cover (1) introduction to background
and potential applications (2%); (2) Review of the state-of-the-art
(3%); (3) Algorithms and critical codes in a nutshell (10%); (4)
Experimental analysis and discussion of proposed methodology (5%).
Reference books:
Math cookbook Linear Algebra Review
Note that:
(1)
mini-projects are not allowed to use any existing toolbox; you have to
write every line codes by yourself.
(2) In final project, you can use the toolbox.
(3) Meanwhile, we will randomly check some students' projects by asking his/her some questions, in order to validate that the projects are done by himself/herself.
Topic | Slides | Exec &Notes | Other material | Web Videos | |
1 | Overview | Introduction | ex1
Notes |
Rcode
|
Intro1 Intro2 |
2 | Linear regression | linear regression | |||
3 | Project -1 |
Oct-13 5:00pm |
project1 |
||
4 | Linear classification |
linear
classification |
ex2 |
||
5 | Linear SVM |
linear_svm |
Chap4(Page170) 6, 7; Chap 9 (Page 368) 1, 2,3 | tutorial: Latex Latex Example Chinese Intro |
|
6 |
SVM |
svm |
|||
7 | project-2 |
deadline: 5:00pm, Nov 19, 2018 | project2 | Naive Bayes | |
8 | neural_network(1) |
nn1 |
NN1 | ||
9 | neural_network(2) | nn2 |
|||
10 | neural_network(3) | nn3 |
|||
11 | Learning theory |
learning
theory |
|||
12 | Project-3 |
deadline: 5:00pm, Jan 5th, 2018 | project3 |
||
13 | Mid-term |
slides |
ex4 |
Notes_Andrew_Ng |
|
14 | Unsupervised Learning |
slides |
EM_GMM |
||
15 | final projects |
|
final_project |
||
16 | semi-supervised learning |
slides |
|||
17 | tree-based method |
slides |
|||
18 |
Good Reading Material: