I have heard people claim (for example in the course Intro to machine learning , lesson 5) that they like the adaboost algorithm without really providing the reason for why.
At the same time, i have briefly gone through the paper Explaining the Success of AdaBoost and Random Forests as Interpolating Classifiers , which suggest that the two are in essence relatively similar.
But in reading about these algorithms i could not find arguments on why to choose one over the other. So my questions are:
- When and why should AdaBoost outperform the random forest?
- What are the other advantages of adaboost over random forest?
- When and why should random forest outperform the adaboost?
- what are the other advantages of random forest over the adaboost?
