Overfitting, Underfitting, and RegularizationMemorizing vs learning; the bias-variance tradeoff; L1, L2, and Elastic Net regularization