Bias_Variance_Tradeoff.md

Bias is the simplifying assumptions made by the model to make the target function easier to approximate.

Variance is the amount that the estimate of the target function will change given different training data.

Trade-off is tension between the error introduced by the bias and the variance

Training set error will tell you about bias, if it is high, bias is generally high

Dev set or test set error will tell you about variance, if it is high then generally variance is high

The above image has high bias because it is mostly linear, but a curved function was needed to fit them well

It is also high variance because it is overfitting on those noises by being too flexible in the middle.