flâneur

boosting, nearest neighbour classification

people.eecs.berkeley.edu · 2,654 words · saved by 1 readers

N/A

Boosting; Nearest Neighbor Classification 147 24 Boosting; Nearest Neighbor Classification ADABOOST (Yoav Freund and Robert Schapire, 1997) AdaBoost (“adaptive boosting”) is an ensemble method for classification (or regression) that – reduces bias [compare with random forests and other ensembles, which reduce variance]; – trains learners on weighted sample points [like bagging]; – uses di↵erent weights for each learner; – increases weights of misclassified training points; – gives bigger votes to more accurate learners. Input: n ⇥ d…

related reading