Regularization
lastmod 2026-07-19created 2015-12-247 edits
Regularization Regularization, in mathematics and statistics and particularly in the fields of machine learning and inverse problems, refers to a process of introducing additional information in order to solve an ill-posed problem or to prevent overfitting.
L1
\[ C = C_0 + \frac{\lambda}{n} \sum_{w} | w | \]
- \(C_0\): the original Cost function.
- \(n\): number of training data.
- \(\lambda\): learning rate
- \(w\): weights
L2
\[ C = C_0 + \frac{\lambda}{2n} \sum_{w} w^2 \]
- \(C_0\): the original Cost function.
- \(n\): number of training data.
- \(\lambda\): learning rate
- \(w\): weights