Bias Variance Tradeoff

In simple words, bias means how far you have come in predicting the desired value from your actual value. It is an approach that can ultimately make or break the model in favor or against your idea. It can also be thought of as a methodological error in your machine learning model due to inappropriate assumptions made by you during the making of that model. Variance is the reverse of bias, It is called the variance when your model performs exceptionally well on the training dataset yet fails to live up to the same standards when running it on an entirely new dataset. The trade-off is simply the tension that exists between the error that is introduced by bias and variance.