5,332
edits
Deep Learning: Difference between revisions
→Kernel Method
| Line 543: | Line 543: | ||
* <math>\phi</math> is fixed! | * <math>\phi</math> is fixed! | ||
* <math>D = O(d^k)</math> | * <math>D = O(d^k)</math> | ||
** For ImageNet, d is approx <math>10^5</math> so <math>D=O(10^15)</math> | ** For ImageNet, <math display="inline">d</math> is approx <math display="inline">10^5</math> so <math display="inline">D=O(10^{15})</math> | ||
;Kernel Trick: | ;Kernel Trick: | ||