1. 주제 : SVM(Support Vector Machine) 세미나
2. 강연자 : 심주용 교수님
3. 일시 : 2013.3.13.(수) 오후 3:00 ~ 4:30
4. 장소 : 경북대학교 IT-4호관 108호
5. 초청교수 : 김일곤 교수님
Support vector machine (SVM), firstly developed by Vapnik (1995, 1998), is being used as a new technique for classification and regression problems. SVM is based on the structural risk minimization (SRM) principle, which has been shown to be superior to traditional empirical risk minimization (ERM) principle.
SRM minimizes an upper bound on the expected risk unlike ERM minimizing the error on the training data. By minimizing this bound, high generalization performance can be achieved.
We give a brief overview of the fundamental principles of SVM.