Glossary term
Glossary term
Training and Fine-Tuning
Synonym for L2 regularization. The term ridge regularization is more frequently used in pure statistics contexts, whereas L2 regularization is used more often in machine learning.
Created for this library
A risk team uses ridge regularization on its linear scorecard to stabilize coefficients when several features are correlated.
A pricing team uses ridge regularization on its demand model so coefficient estimates stay stable across vintages of training data.
A research team uses ridge regularization as the default linear-model regularizer when no strong sparsity is needed.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License