Bias-Variance Tradeoff

What is Bias-Variance Tradeoff?

The conflict in minimizing two sources of error: 'Bias' (error from oversimplifying the model) and 'Variance' (error from excessive sensitivity to small fluctuations in the training set). A good model must find the sweet spot between underfitting (high bias) and overfitting (high variance).

Where did the term "Bias-Variance Tradeoff" come from?

A core theorem in statistical learning theory.

How is "Bias-Variance Tradeoff" used today?

Central to understanding model performance; it explains why more complex models aren't always better.

Related Terms