Fridge regression is a new penalty family that keeps a prespecified number of coefficients non-zero under extreme regularization, combining lasso speed with best-subset interpretability.
D., Joulin, A., Bach, F
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ME 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Non-null Shrinkage Regression and Subset Selection via the Fractional Ridge Regression
Fridge regression is a new penalty family that keeps a prespecified number of coefficients non-zero under extreme regularization, combining lasso speed with best-subset interpretability.