Revision history [back]

Hmm there is a big difference between AdaBoost and Neural Networks... so I would not say that they are the same. Also the stumps is a default value, but you can take whatever structure of a tree as weak classifier, so the limitation and buildup is different. Besides, a cascade enables to use the early rejection principle, while a neural net needs to process each input completely and thus makes in computationally heavy for a sliding window approach.