I've done no experiment, but I predict a boosted forest would perform poorly here. Since the first tree learns the majority of the information (and each of the

I've done no experiment, but I predict a boosted forest would perform poorly here. Since the first tree learns the majority of the information (and each of the following trees only fits the errors, and are therefore less "meaningful"). Since the kernel approach suggested here weights all the trees equally, the results probably won't be as good. Maybe if you gave the first trees more weight, though...