2013 Fall Undergraduate Course in
Computer Science and Engineering
Instructor:
Prof. Byoung-Tak Zhang
TA:
Sang-Woo Lee, Hyo-Sun Chun
Classroom:
302-209
Time:
Tue & Thu, 11:00-12:15
Reference:
- [1] Bishop, C. M., Pattern Recognition and Machine Learning, Springer-Verlag, 2006.
- A Tutorial on Learning With Bayesian Networks (11¿ù 27ÀÏ¿¡ º¯°æ)
Evaluation:
- Two open-book exams (50%)
- Two miniprojects + two reports + one poster presentation (40%)
- Participation and discussion (10%)
Announcement:
- Áú¹®°ú Åä·ÐÀ» À§ÇÑ °Ô½ÃÆÇÀÌ »ý¼ºµÇ¾ú½À´Ï´Ù.
- 9/26 (¸ñ)Àº ÈÞ°ÀÔ´Ï´Ù. (°ø´ë ÃàÁ¦ - °ø¸í)
- ¸Å ¼÷Á¦ÀÇ deadlineÀº ¼÷Á¦°¡ °øÁö µÈ ½Ç½À ½Ã°£À¸·Î ºÎÅÍ 1ÁÖÀÏ µÚ ÀÚÁ¤ÀÔ´Ï´Ù.
- ´Ü, ¼÷Á¦2ÀÇ ±âÇÑÀº 9/24 (È) ÀÚÁ¤±îÁöÀÔ´Ï´Ù. (toolbox ÇÔ¼ö ¾øÀÌ Äڵ尡 µ¹¾Æ¾ß ÇÕ´Ï´Ù.)
- Áß°£ ÇÁ·ÎÁ§Æ®°¡ °øÁöµÇ¾ú½À´Ï´Ù. ±âÇÑÀº 10/10 (¸ñ) ÀÚÁ¤±îÁöÀÔ´Ï´Ù.
- 10/4 (±Ý) 4:00~6:00 ¿¡ º¸°ÀÌ ÀÖ½À´Ï´Ù.
- Ãâ¼® ¹× ¼÷Á¦ ÇöÀç Á¡¼ö°¡ °øÁöµÇ¾ú½À´Ï´Ù. ÀÌ»óÇÑ Á¡ÀÌ ÀÖÀ¸½Å ºÐµéÀº ¹®ÀÇÇØÁֽñ⠹ٶø´Ï´Ù.
- 10/15 È¿äÀÏ ¼ö¾÷Àº °³±³±â³äÀÏ·Î ÀÎÇØ ÈÞ°ÇÕ´Ï´Ù.
- ´ÙÀ½ ³í¹®À» ÀÐÀ¸½Ã±â ¹Ù¶ø´Ï´Ù. Next-Generation Machine Learning Technologies
- ±â¸» ÇÁ·ÎÁ§Æ®°¡ °øÁöµÇ¾ú½À´Ï´Ù.
(µ¥ÀÌÅÍ,
Âü°í³í¹®,
µ¥¸ðÄÚµå)
DEADLINE: 11/28 (¸ñ) ÀÚÁ¤ => µ¥µå¶óÀÎ ÀÌÈÄ¿¡µµ ±â¸»°í»çÀϱîÁö °è¼Ó ¹ÞÀ¸³ª, ¾à°£ÀÇ °¨Á¡ÀÌ ÀÖ½À´Ï´Ù
- 11/14ÀÏ, 19ÀÏ Æ÷½ºÅÍ ¹ßÇ¥°¡ ¿¹Á¤µÇ¾î ÀÖ½À´Ï´Ù. (ÁÖÁ¦3,4¸ñ·Ïº¸±â)
Æ÷½ºÅÍ ¹ßÇ¥ Àå¼Ò: 302µ¿ 308È£ / 14ÀÏ-ÁÖÁ¦1 / 19ÀÏ-ÁÖÁ¦2, 3, 4
Æ÷½ºÅÍ ÀÚ·áÀ» Á¾° Àü±îÁö hschun@bi.snu.ac.kr·Î Á¦Ãâ¹Ù¶ø´Ï´Ù
- Áß°£°í»ç Á¡¼ö°¡ °øÁöµÇ¾ú½À´Ï´Ù. Claim±â°£Àº 11/18 ~ 11/23 ÀÌ¸ç »çÀü¿¡ slee at bi.snu.ac.kr·Î ¹®ÀÇÇØÁֽñ⠹ٶø´Ï´Ù.
- ±â¸»°í»ç´Â 12¿ù 17ÀÏ È¿äÀÏÀÔ´Ï´Ù
- ¼öÁ¤µÈ Áß°£°í»ç Á¡¼ö¿Í ±â¸»°í»ç Á¡¼ö°¡ °øÁöµÇ¾ú½À´Ï´Ù. ½ÃÇè °á°úÀÇ Claim±â°£Àº 12/26 ÀÌ¸ç »çÀü¿¡ ¹Ýµå½Ã slee at bi.snu.ac.kr·Î ¹®ÀÇÇØÁֽñ⠹ٶø´Ï´Ù.
- ±â¸» ½Ç½À ¹× ÇÁ·ÎÁ§Æ® Á¡¼ö°¡ °øÁöµÇ¾ú½À´Ï´Ù. ½Ç½À °á°ú¿¹»ó°ú Å©°Ô ´Ù¸¥ Á¡¼ö°¡ ³ª¿À½Å ºÐµéÀº slee at bi.snu.ac.kr·Î ¹®ÀÇÇØÁֽñ⠹ٶø´Ï´Ù.
Course Description:
- Machine learning studies learning systems, i.e. agents that improve their performance automatically by acquiring knowledge through experience and data collected from interaction with their environments.
Machine learning plays a key role in modern artificial intelligence and is applied to a wide range of practical applications, including data mining, information retrieval, natural language processing, computer vision, robotics, social networks, and mobile services.
The objective of this course is two-fold. One is to provide the students with the principles and mathematical tools for machine learning.
Another is to introduce the architectures and algorithms for representative models of machine learning. These include pattern classification models, kernel machines, clustering algorithms, and probabilistic graphical models.
|
-
Week |
Topics |
Slides |
Week 1
(9/3, 9/5) |
- What's Machine Learning and Why?
- Bayesian inference and information theory
|
Pictures of Blackboard (9/5)
1
2
3
4
|
Week 2
(9/10, 9/12) |
- Classification practice 1
- Classification practice 2
|
Matlab Tutorial 1
Problem 1
MNIST Dataset
Matlab Tutorial 2
Problem 2
|
|
Week 3
(9/17) |
|
Pictures of Blackboard (9/17)
5
6
7
8
9
10
|
|
Week 4
(9/24) |
- Kernel machine practice 1
|
Problem 3
Radial Basis Function Networks
|
|
Week 5
(10/1) |
|
Pictures of Blackboard (10/1)
11
12
13
14
Pictures of Blackboard (10/4)
15
16
17
18
19
|
Week 6
(10/8, 10/10) |
- Kernel machine theory 1
- Kernel machine theory 2
- Due of first project report (10/10)
|
Pictures of Blackboard (10/8)
20
21
22
Pictures of Blackboard (10/10)
23
24
25
26
|
Week 7
(10/15, 10/17) |
- 10/15 °³±³±â³äÀÏ ÈÞ°
- Clustering theory 1
|
K-means Clustering
Pictures of Blackboard (10/17)
27
28
29
|
Week 8
(10/22, 10/24) |
- Clustering theory 2
- Clustering theory 3
|
Pictures of Blackboard (10/22)
30
31
32
33
34
35
Pictures of Blackboard (10/24)
36
37
38
|
|
Week 9
(10/29, 10/31) |
- Exam 1
- Clustering practice 1
|
|
|
Week 10
(11/5, 11/7) |
- Clustering practice 2
- Clustering practice 3
|
Problem 5
kmeans_skeleton
practice_image
(small)
Problem 6
GMM_skeleton,
mvnpdf.m
|
|
Week 11
(11/12, 11/14) |
- Project practice
- Poster presentation 1
|
Æ÷½ºÅÍ ¾È³» |
|
Week 12
(11/19, 11/21) |
- Poster presentation 2
- Probabilistic graphical models 1
|
Bayesian Networks
BN reference link |
|
Week 13
(11/26, 11/28) |
- Probabilistic graphical models 2
- Probabilistic graphical models 3
|
Markov Random Fields
Deep Belief Networks(Âü°í)
Pictures of Blackboard (11/28)
39
40
41
42
|
|
Week 14
(12/3) |
|
|
|
Week 16
(12/17) |
|
|
|
|