Difference between revisions of "TDSM 11.3"
From The Data Science Design Manual Wikia
Caitaozhan (talk | contribs) m (regularization) |
m |
||
| Line 2: | Line 2: | ||
Textbook page 286: regularization is the trick of adding secondary terms to the objective function to favor models that keep coefficient small. | Textbook page 286: regularization is the trick of adding secondary terms to the objective function to favor models that keep coefficient small. | ||
| + | |||
| + | Edit: It punishes larger coefficients | ||
Latest revision as of 00:05, 13 December 2017
Regularization is a technique used in an attempt to solve the overfitting problem in statistical models. (from Wikipedia)
Textbook page 286: regularization is the trick of adding secondary terms to the objective function to favor models that keep coefficient small.
Edit: It punishes larger coefficients
