A desicion-theoretic generalization of on-line learning and an application to boosting Export

@incollection{citeulike:3457471, abstract = {We consider the problem of dynamically apportioning resources among a set of options in a worst-case on-line framework. The model we study can be interpreted as a broad, abstract extension of the well-studied on-line prediction model to a general decision-theoretic setting. We show that the multiplicative weight-update rule of Littlestone and Warmuth [10] can be adapted to this mode yielding bounds that are slightly weaker in some cases, but applicable to a considerably more general class of learning problems. We show how the resulting learning algorithm can be applied to a variety of problems, including gambling, multiple-outcome prediction, repeated games and prediction of points in n . We also show how the weight-update rule can be used to derive a new boosting algorithm which does not require prior knowledge about the performance of the weak learning algorithm.}, author = {Freund, Yoav and Schapire, Robert}, citeulike-article-id = {3457471}, citeulike-linkout-0 = {http://dx.doi.org/10.1007/3-540-59119-2_166}, citeulike-linkout-1 = {http://www.springerlink.com/content/7qp818392282l161}, doi = {10.1007/3-540-59119-2_166}, journal = {Computational Learning Theory}, keywords = {adaboost}, pages = {23--37}, posted-at = {2008-10-28 13:57:49}, priority = {2}, title = {A desicion-theoretic generalization of on-line learning and an application to boosting}, url = {http://dx.doi.org/10.1007/3-540-59119-2_166}, year = {1995} }

TY - CHAP ID - citeulike:3457471 L3 - citeulike-article-id:3457471 N2 - We consider the problem of dynamically apportioning resources among a set of options in a worst-case on-line framework. The model we study can be interpreted as a broad, abstract extension of the well-studied on-line prediction model to a general decision-theoretic setting. We show that the multiplicative weight-update rule of Littlestone and Warmuth [10] can be adapted to this mode yielding bounds that are slightly weaker in some cases, but applicable to a considerably more general class of learning problems. We show how the resulting learning algorithm can be applied to a variety of problems, including gambling, multiple-outcome prediction, repeated games and prediction of points in n . We also show how the weight-update rule can be used to derive a new boosting algorithm which does not require prior knowledge about the performance of the weak learning algorithm. JF - Computational Learning Theory EP - 37 TI - A desicion-theoretic generalization of on-line learning and an application to boosting SP - 23 KW - adaboost AU - Freund, Yoav AU - Schapire, Robert PY - 1995/// UR - http://dx.doi.org/10.1007/3-540-59119-2_166 ER -