Estimate of error bounds in the improved support vector regression

Abstract

An estimate of a generalization error bound of the improved support vector regression (SVR) is provided based on our previous work. The boundedness of the error of the improved SVR is proved when the algorithm is applied to the function approximation.

Keywords:

• support vector regression (SVR)
• generalization error
• approximation error
• estimation error

Supported by the Science-Technology Development Project of Jilin Province of China (Grant No. 20030520), the Key Project of Ministry of Education of China (Grant No. 02090) and the National Natural Science Foundation of China (Grant No. 19901009)

Supported by the Science-Technology Development Project of Jilin Province of China (Grant No. 20030520), the Key Project of Ministry of Education of China (Grant No. 02090) and the National Natural Science Foundation of China (Grant No. 19901009)

Notes

Supported by the Science-Technology Development Project of Jilin Province of China (Grant No. 20030520), the Key Project of Ministry of Education of China (Grant No. 02090) and the National Natural Science Foundation of China (Grant No. 19901009)