Statistical learning theory and state of the art in SVM

Statistical learning theory started more than 30 years ago. Until the middle of the 1990´s, the success of support vector machine (SVM) in solving real-life problems made it not only a tool for the theoretical analysis but also a tool for creating practical algorithms for real-world problems. In this paper, we present a general overview of statistical learning theory and theoretically analyze the reason of overfitting problem in statistical learning. We also describe the current state of the art in SVM. Finally, as an application of SVM, we present experimental results in our implementation of SVM and demonstrate its advantage in multiuser detection problem.