Learning the hypotheses space from data through a U-curve algorithm: a statistically consistent complexity regularizer for Model Selection

Open in new window