Presentation | 2007-12-22 Optimizing SVR Hyperparameters via Fast Cross-Validation using AOSVR Masayuki KARASUYAMA, Ryohei NAKANO, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The performance of Support Vector Regression (SVR) deeply depends on its hyperparameters. such as an insensitive zone thickness, a penalty factor, and kernel parameters. A method called MCV-SVR was once proposed, which optimizes SVR hyperparameters so that cross-validation error is minimized. However, the computational cost of CV is usually high. In this paper we apply Accurate Online Support Vector Regression (AOSVR) to the MCV-SVR cross-validation procedure. The AOSVR enables an efficient update of a trained SVR function. We show the AOSVR dramatically accelerates the MCV-SVR. Moreover, our experiments showed our faster MCV-SVR has better generalization than other existing methods. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | support vector machines / support vector regression / mimum cross-validation |
Paper # | NC2007-73 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 2007/12/15(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Neurocomputing (NC) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Optimizing SVR Hyperparameters via Fast Cross-Validation using AOSVR |
Sub Title (in English) | |
Keyword(1) | support vector machines |
Keyword(2) | support vector regression |
Keyword(3) | mimum cross-validation |
1st Author's Name | Masayuki KARASUYAMA |
1st Author's Affiliation | Nagoya Institute of Technology() |
2nd Author's Name | Ryohei NAKANO |
2nd Author's Affiliation | Nagoya Institute of Technology |
Date | 2007-12-22 |
Paper # | NC2007-73 |
Volume (vol) | vol.107 |
Number (no) | 410 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |