Research Article
A Faster Gradient Ascent Learning Algorithm for Nonlinear SVM
Table 1
The comparative analysis in terms of the number of iterations and the values of
.
| | Number of iterations/the estimation of the maximum value of (4) | SMO | The variant of the gradient algorithm using the weight parameter (15) | The variant of the gradient algorithm using the weight parameter (20) | The variant of the gradient algorithm using the weight parameter (24) |
| 0.15 | 712/139.9 | 947/139.84 | 994/139.82 | 901/139.85 | 0.3 | 661/71.37 | 458/71.12 | 494/71.11 | 437/71.12 | 0.5 | 322/43.96 | 269/43.72 | 300/43.72 | 262/43.72 | 0.8 | 312/28.64 | 180/28.38 | 195/28.38 | 180/28.38 | 1 | 319/23.63 | 143/23.43 | 157/23.42 | 143/23.43 | 1.2 | 311/20.34 | 113/20.17 | 133/20.17 | 117/20.17 | 1.4 | 279/18.05 | 88/17.9 | 116/17.9 | 104/17.9 |
|
|