Glossary term
Glossary term
Foundations
Any mechanism that reduces overfitting. Popular types of regularization include:
early stopping (this is not a formal regularization method, but can effectively limit overfitting)
Regularization can also be defined as the penalty on a model's complexity.
Click the icon for additional notes.
See Overfitting: Model complexity in Machine Learning Crash Course for more information.
Created for this library
An ML team uses L2 regularization on its production scorecard to stabilize coefficients across retrain vintages.
A research team applies dropout regularization in its transformer blocks to keep generalization stable on small datasets.
A risk modeling team uses regularization to control overfitting on small datasets typical of niche credit products.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License