Softmax Regression
From Ufldl
m (→Binary Logistic Regression) |
(→Parameterization) |
||
Line 171: | Line 171: | ||
Showing that only <math>n-1</math> parameters are required. | Showing that only <math>n-1</math> parameters are required. | ||
- | In practice, however, it is often easier to implement the version which is over-parametrized although both methods will lead to the same classifier. | + | In practice, however, it is often easier to implement the version which is over-parametrized although both methods will lead to approximately the same classifier. |
=== Binary Logistic Regression === | === Binary Logistic Regression === |