Summary: A NEW SYSTEM IDENTIFICATION METHOD
BASED ON SUPPORT VECTOR MACHINES
Shuichi Adachi and Tomonori Ogawa
Department of Electrical and Electronic Engineering
7-1-2 Yoto, Utsunomiya, 321-8585, Japan
Phone/Fax : +81 28 689 6125
e-mail : firstname.lastname@example.org
Abstract: Support Vector Machines (SVM) have become a subject of intensive study in sta-
tistical learning theory. They have been applied to successfully to classification problems and
recently extended to regression problems. Support vector machines for regression problems is
called Support Vector Regression (SVR). In this paper, a brief introduction to SVR is presented
and then a new system identification method based on SVR is proposed for linear in parameter
models. The effectiveness of the proposed method is examined through numerical examples.
Keywords: system identification, statistical learning theory, support vector machines, regular-
ization, robust estimation.
Support Vector Machines (SVM) proposed by V.N.
Vapnik early in 90's have become a subject of inten-
sive study in statistical learning theory. They have