This is a project for me to understand:
- the difference between Random Forest and Extremely Randomized Trees (ERT)
- the effect of the differences on the bias and variance
- the advantages of ERT's
It also contains a notebook that shows the effect of various machine-learning model parameters on the bias and variance part of the error. As it turns out the effect of model parameters on bias and variance is almost always explained by model complexity. More complexity allows for a smaller bias (average prediction error for multiple independent runs) and higher variance (differences in prediction error for multiple independent runs).
Geurts, P., Ernst, D., & Wehenkel, L. (2006). Extremely randomized trees. Machine Learning, 63(1), 3–42. https://doi.org/10.1007/s10994-006-6226-1