Bayes Error
Bayes error is the minimum classification error you can have for the true distributions of the data. Regardless of how big your dataset is, you can’t get lower than this error. It is obtained by setting the decision boundarydecision boundary at the point where two models’ probability distributions intersect.
