https://scholars.lib.ntu.edu.tw/handle/123456789/115492
Title: | Leave-one-out Bounds for Support Vector Regression Model Selection | Authors: | Chang, Ming-Wei Lin, Chih-Jen |
Issue Date: | 2005 | Start page/Pages: | 1188-1222 | Source: | Neural Computation | Abstract: | Minimizing bounds of leave-one-out (loo) errors is an important & efficient approach for support vector machine (SVM) model selection. Past research focuses on their use for classification but not regression. In this article, we derive various loo bounds for support vector regression (SVR) & discuss the difference from those for classification. Experiments demonstrate that the proposed bounds are competitive with Bayesian SVR for parameter selection. We also discuss the differentiability of loo bounds. |
URI: | http://ntur.lib.ntu.edu.tw//handle/246246/20060927122856648470 | Other Identifiers: | 20060927122856648470 | DOI: | 10.1162/0899766053491869 |
Appears in Collections: | 資訊工程學系 |
File | Description | Size | Format | |
---|---|---|---|---|
svrbound.pdf | 250.16 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.