The words contained in this file might help you see if this file matches what you are looking for:

...Csc fall machine learning data mining ensemble methods slides by rich zemel typical application classi ication of iers is a set whose individual decisions combined in some way to classify new examples simplest approach generate multiple each votes on test instance take majority as different due sampling training or randomized parameters within the algorithm aim simple mediocre and transform it into super ier without requiring any fancy summary differ strategy combination method parallel with sets bagging sequential iteratively re weighting so current focuses hard boosting objective encouraging division labor mixture experts notes also known meta typically applied weak models such decision stumps single node trees linear variance bias tradeo minimize two errors error from sensitivity small luctuations erroneous assumptions model decomposition analyzing generalization sum terms irreducible resulting problem itself...