Softmax Regression
From Ufldl
| Line 160: | Line 160: | ||
| - | + | == Properties of softmax regression parameterization == | |
Softmax regression has an unusual property that it has "too many," or "redundant", parameters. Concretely, | Softmax regression has an unusual property that it has "too many," or "redundant", parameters. Concretely, | ||
| Line 210: | Line 210: | ||
| - | + | == Weight Decay == | |
<!-- | <!-- | ||
| Line 282: | Line 282: | ||
| - | == | + | == Relationship to Logistic Regression == |
In the special case where <math>k = 2</math>, one can also show that softmax regression reduces to logistic regression. | In the special case where <math>k = 2</math>, one can also show that softmax regression reduces to logistic regression. | ||
| Line 336: | Line 336: | ||
| - | + | == Softmax Regression vs. k Binary Classifiers == | |
Suppose you are working on a music classification application, and there are | Suppose you are working on a music classification application, and there are | ||
