| Statistical Learning Workshop研討會
林智仁 助理教授
(
台灣大學資訊工程學系 )
Support Vector Machines: Introduction and Theory.
摘要
| | 摘 要
Support vector machine (SVM) is a new and promising technique for data classification and regression. This talk will include:
1.Basic concepts of SVM.
We will discuss linear and nonlinear support vector classification. The relation with neural network will also briefed mentioned. The support vector regression will also be introduced.
2.The learning theory behind SVM.
Mainly we will discuss some VC bounds which are directly related to SVM.
3.Some practical applications of SVM:
We will discuss one practical example for time series prediction. The tool used is support vector regression.
4.(if time allowed) Algorithms and implementations of SVM.
We will demonstrate our research on designing and implementing SVM algorithms and software. In particular we discuss what an effective SVM software should be like. For example, issues such as model selection, handling unbalanced data, user interface, etc
|
91年1月5日 (星期六)
AM9:00-12:00
台灣大學數學系新數館308室
|