- Lecturer
- Yamashita Yukihiko
- Place
- Thr7-8(S611)
- Credits
- Lecture2 Exercise0 Experiment0
- Code
- 70042
- Syllabus updated
- 2015/2/12
- Lecture notes updated
- 2015/2/12
- Semester
- Fall Semester
Outline of lecture
Please take the presentation slide of this lecture from
http://www.ide.titech.ac.jp/~yamasita/MMS/
Linear algebra (eigenvalue problem, singular value decomposition, generalized inverse matrix),
statistics (estimation and test),
and optimization technique (gradient method, conjugate gradient method, and quasi-Newton method) are lectured.
Purpose of lecture
The objective of this course is to provide fundamental optimization technique
and statistics to handle various quantities with respect to international development.
In order to understand those knowledges, basic mathematics for them is also provided.
Plan of lecture
1. Introduction
2. Eigenvalue decomposition and singular value decomposition
3. Generalized inverses of matrix
4. Probability (Definition of Random variable)
5. Estimator (Maximum likelihood estimator and Bayesian estimator)
6. Cramer-Rao lower bound
7. Principle component analysis
8. Regression
9. Testing
10. Statistical learning theory
11. Maximum gradient method and Conjugate gradient method
12. Newton method and Quasi-Newton method
13. Lagrange窶冱 method and Karush–Kuhn–Tucker condition
14. Dual problem
15. Penalty method
Textbook and reference
Copy of slide is prepared
Related and/or prerequisite courses
Linear algebra
Statistics
Evaluation
Attendance
Homework
Report
