This flashcard is just one of a free flashcard set. See all flashcards!

Bias is how accurate a model is across different training sets (how general a model is).

Variance is how sensitive a model is to small changes in the training set.

Error = Variance + Bias^2 + Noise
We want to minimize the bias and the variance of the model error.
High bias -> underfitting (model too simple)
High variance -> overfitting (model too complex)

To achieve good performance on data outside the training set a tradeoff must be made.