Softmax Regression
From Ufldl
(→Optimizing Softmax Regression) |
|||
Line 54: | Line 54: | ||
</math> | </math> | ||
- | Unfortunately, there is no | + | Unfortunately, there is no closed form solution to this optimization problem (although it is concave), and we usually use an off-the-shelf optimization method (e.g., L-BFGS, stochastic gradient descent) to find the optimal parameters. Using these optimization methods require computing the gradient (<math>\ell(\theta)</math> w.r.t. <math>\theta_{k}</math>), which can can be derived as follows: |
<math> | <math> |