Monthly
288 pp. per issue
6 x 9, illustrated
ISSN
0899-7667
E-ISSN
1530-888X
2014 Impact factor:
2.21

Neural Computation

August 1, 2002, Vol. 14, No. 8, Pages 1959-1977
(doi: 10.1162/089976602760128081)
© 2002 Massachusetts Institute of Technology
Training v-Support Vector Regression: Theory and Algorithms
Article PDF (302.11 KB)
Abstract

We discuss the relation between ϵ-support vector regression (ϵ-SVR) and v-support vector regression (v-SVR). In particular, we focus on properties that are different from those of C-support vector classification (C-SVC) and v-support vector classification (v-SVC). We then discuss some issues that do not occur in the case of classification: the possible range of ϵ and the scaling of target values. A practical decomposition method for v-SVR is implemented, and computational experiments are conducted. We show some interesting numerical observations specific to regression.