인공지능 특강 (CSI8751-01)
Topics in AI:
 Machine Learning: Methodologies and Applications

---

◈ Teaching Staffs

• Lecturer
• Cho, Sung-Bae (C515; 2123-2720, sbcho@cs.yonsei.ac.kr)
• Web page: http://sclab.yonsei.ac.kr/Courses/10TAI
• Class hours: (Eng, A019) Tue 5, Thu 5,6
• Office hours: Tue 7,8
• TA: Lee, Young-Seol (2123-4803, tiras@sclab.yonsei.ac.kr)

---

◈ Course Objectives

• Understanding machine learning technologies such as decision tree, artificial neural networks, genetic algorithms, etc
• Developing systems to solve complex real-world problems effectively by applying them

---

◈ Textbook

• Textbook
• T.M. Mitchell, Machine Learning, McGraw Hill, 1997
• References
• T. Dean, J. Allen and Y. Aloimonos, Artificial Intelligence: Theory and Practice, The Benjamin/Cummings Pub., 1995
• S. Russell and P. Norvig, Artificial Intelligence: A Modern Approach, Prentice Hall, 1995
• P.H. Winston, Artificial Intelligence, 3rd Ed, Addison Wesley, 1992
• P. Baldi, Bioinformatics: The Machine Learning Approach, MIT Press, 2001

---

◈ Evaluation Criteria

• Term Project (written report and an oral presentation)        : 40%

ü Related material ("When Semi-Supervised Learning Meets Ensemble Learning")

• Final Exam                                                                       : 20%
• Homeworks                                                               : 20%
1. HW#1(Report for Weka)
2. HW#2(Report for Decision Tree C4.5 or J48)

(1) measure to select attributes (결정 트리를 구성할 때 가장 적합한 root node 를 선택하는 기준에 대해서 설명)

(2) stopping condition (결정 트리를 구성할 때 지나치게 trivial 한 attribute 까지 모두 고려하지 않도록 적정 수준에서 멈추기 위한 조건에 대해서 설명)

(3) solving problem using sample data (37 종류의 데이터 가운데 자기에게 할당된 데이터에 대해서 Weka 를 테스트할 것)

3. HW#3

Essay about bias-variance dillema after reading Geman's neural computation journal paper (1992)

(2~3 pages)

• Presentation & Participaption                                              : 20%

◈ Term Project (Oral presentation is required) :

• Theoretical Issue (Analysis, Experiment, Simulation) : Originality
• Interesting Programming (Game, Demo, etc) : Performance
• Survey : Completeness
• List of Possible Projects
• Tangible Agent
• Integrated Model
• Life Browser
• Bayesian Network for Middleware
• Cluster GA
• SASOM for Motion Recognition
• Evolvability
• Evolutionary Neural Networks

---

◈ Schedule

l 11/9 (No class because of WCU meeting)

Week	Details
9/2	Course Introduction
9/7, 9/9	Introduction (Mitchell, Ch1)
9/14, 9/16	A Concept Learning (Mitchell, Ch2)
9/21, 9/23	HW#1 (Chu-Seok)
9/28, 9/30	Decision Tree Learning (Mitchell, Ch3)
10/5, 10/7	Artificial Neural Networks (Mitchell, Ch4)
10/12, 10/14	Evaluating Hypothesis (Mitchell, Ch5)
10/19, 10/21	Term-paper proposal & (Talk : Multi-objective Optimization Using Particle Swarm Optimization)
10/26, 10/28	Bayesian Learning (Mitchell, Ch6)
11/2, 11/4	Computational Learning Theory (Mitchell, Ch7)
11/9, 11/11	11/9(No class because of WCU meeting) 11/10 (Extra class 16:00~18:00 - room: C029 )
11/16, 11/18	Instance-based Learning (Mitchell, Ch8) Project progress presentation (M. Arshad Awan , Dong-Kyu Jeon)
11/23, 11/25	Genetic Algorithms (Mitchell, Ch9) Project progress presentation (Jong-won Yoon)
11/30, 12/2	Final Exam
12/7, 12/9	Final presentation
12/14, 12/16	Due date for term-paper