https://scholars.lib.ntu.edu.tw/handle/123456789/115213
標題: | Decomposition Methods for Linear Support Vector Machines | 作者: | Kao, Wei-Chun Chung, Kai-Min Sun, Chia-Liang Lin, Chih-Jen |
公開日期: | 2004 | 出版社: | 臺北市:國立臺灣大學資訊工程學系 | 摘要: | In this paper, we show that decomposition methods with alpha seeding are extremely useful for solving a sequence of linear SVMs with more data than attributes. This strategy is motivated from (Keerthi and Lin 2003) which proved that for an SVM with data not linearly separable, after C is large enough, the dual solutions are at the same face. We explain why a direct use of decomposition methods for linear SVMs is sometimes very slow and then analyze why alpha seeding is much more effective for linear than nonlinear SVMs. We also conduct comparisons with other methods which are efficient for linear SVMs, and demonstrate the effectiveness of alpha seeding techniques for helping the model selection. |
URI: | http://ntur.lib.ntu.edu.tw//handle/246246/20060927122900179258 | 其他識別: | 20060927122900179258 |
顯示於: | 資訊工程學系 |
檔案 | 描述 | 大小 | 格式 | |
---|---|---|---|---|
linear.pdf | 214.18 kB | Adobe PDF | 檢視/開啟 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。