2015

Please see this file.

Course | Instructor | Courseware | Keynote | |
---|---|---|---|---|

March, 2015 | Multivariate Analysis | Hongxin Zhang | see below ... | 1 2 3 4 |

April, 2015 | Nonlinear Optimization | Hongxin Zhang | see below ... | 5 6 7 8 |

The course is open to Ph.D students of College of Computer Science and graduate students of related majors, Zhejiang University.

Tuesday, 18:30-21:30. Cao Guang Biao Building West 2-202, Yu Quan District, Zhejiang University.

**Deadline**: Please hand out your **1** course paper and **5** selected homework (or exercises) by 2015-06-01.

- Homework 01 ⇒ curve fitting
- Homework 02 ⇒ PCA
- Homework 03 ⇒ 2D MOG and k-means
- Homework 04 ⇒ L-M algorithm
- Homework 05 ⇒ 2D SVM

More optional excercises:

- Excercise 01 ⇒ Gaussian distribution and its conjugate prior
- Excercise 02 ⇒ kernel PCA
- Excercise 03 ⇒ RPCA
- Excercise 04 ⇒ ISOMAP and LLE
- …

The study of learning from data is commercially and scientifically important. This one month short course is designed to give first year Ph.D. students a thorough grounding in the methodologies, technologies, mathematics and algorithms currently needed by people who do research in learning and data mining or who may need to apply learning or data mining techniques to a target problem. The topics of the course draw from classical statistics, from machine learning, from data mining, from Bayesian statistics and from statistical algorithmics.

Students entering the class should have a pre-existing working knowledge of probability, statistics and algorithms, though the class has been designed to allow students with a strong numerate background to catch up and fully participate.

Topic | Date | Slides | Homework |
---|---|---|---|

Introduction | 2015.03.10 | Introduction | HW01 |

Why data driven | |||

Point estimation | |||

Component Analysis | 2015.03.17 | PCA and its related techniques | HW02 |

Distance and similarity | 2015.03.24 | Distance, similarity Clustering | HW03 |

Graphical Models | 2015.03.31 | Graphical Models Hidden Markov Models |

Optimization methods, both linear and non-linear ones, are important mathematical techniques for computer science. This one month short course is designed to give first year Ph.D. students a thorough grounding in the methodologies, technologies, mathematics and algorithms currently needed by people who are doing research related to linear and non-linear optimization. The topics of the course draw mainly from linear programming, quadratic programming and nonlinear optimization.

Students entering the class should have a pre-existing working knowledge of fundamental mathematics and algorithms, though the class has been designed to allow students with a strong numerate background to catch up and fully participate.

Topic | Date | Slides | Homework |
---|---|---|---|

Linear programming | 2015.04.07 | Linear programming and simplex methods | |

Linear programming | 2015.04.14 | Dual methods | |

SVM | |||

Non-linear optimization | 2015.04.21 | NP | HW04 |

Quadratic programming | 2015.04.28 | QP | HW05 |

- 袁亚湘，孙文瑜. 最优化理论与方法，科学出版社.
- 张建中，许绍吉. 线性规划. 科学出版社.
- 黄红选，韩继业. 数学规划. 清华大学出版社.
- Stephen Boyd and Lieven Vandenberghe. Convex Optimization. Cambridge University Press.

Python is a powerful but easy-to-use script language for daily software development. In this course, we mainly use it as a standard training tool to temper mathematical skills. Several useful links are listed as following for your reference.

Learning Python:

- An active Chinese python forum: http://wiki.woodpecker.org.cn. You can find a Chinese tutorial book in this website. Alternatively, you can read the following simplified course in Chinese http://woodpecker.org.cn/abyteofpython_cn/chinese/.
- A Beginner's Python Tutorial in English

Most useful Python packages in this course:

- Plot the results in matplotlib and gnuplot
- NumPy is the fundamental package needed for scientific computing with Python.

Other related modules:

2015.txt · Last modified: 2015/04/29 11:39 by hongxin

Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Noncommercial-Share Alike 4.0 International