Bias-Variance Tradeoff
A predictive model bias–variance tradeoff indicates that lower bias can lead to higher variance and vice versa.
Bias-variance tradeoff is related to model data fitting biased underfitting and variance overfitting.
During model training, the best overall accuracy is achieved near the area where the bias and variance curves cross.
The tradeoff can be seen in this simple equation:
In some Machine Learning models, such as Artificial Neural Networks, it’s sometimes possible to modify the model in a way that improves both bias and variance. In the case of Artificial Neural Networks, increasing the number of graph layers and/or nodes can improve both factors.