Concepts2

Groups

Data Augmentation Theory

Data augmentation expands the training distribution by applying label-preserving transformations to inputs, which lowers overfitting and improves generalization.

#data augmentation#vicinal risk minimization#invariance+12

📚TheoryIntermediate

Bias-Variance Tradeoff

The bias–variance tradeoff explains how prediction error splits into bias squared, variance, and irreducible noise for squared loss.

#bias variance tradeoff