Problem
You receive a proposal to combine 200 weak leaners, where each learner would predict voices as female (i.e., y=0). You have learned that the weak learner has accuracy of around 51% for both training and test set. Do you think the proposal is reasonable? Explain briefly.