159 search results - page 1 / 32 » Geometric Bounds for Generalization in Boosting |
COLT
2001
Springer
13 years 8 months ago
2001
Springer
We consider geometric conditions on a labeled data set which guarantee that boosting algorithms work well when linear classifiers are used as weak learners. We start by providing ...
ML
2002
ACM
13 years 4 months ago
2002
ACM
We consider the existence of a linear weak learner for boosting algorithms. A weak learner for binary classification problems is required to achieve a weighted empirical error on t...
JMLR
2006
13 years 4 months ago
2006
We give a review of various aspects of boosting, clarifying the issues through a few simple results, and relate our work and that of others to the minimax paradigm of statistics. ...
FOCS
2006
IEEE
13 years 10 months ago
2006
IEEE
We prove an upper bound, tight up to a factor of 2, for the number of vertices of level at most in an arrangement of n halfspaces in Rd , for arbitrary n and d (in particular, the...
EUROCOLT
1995
Springer
13 years 7 months ago
1995
Springer
k. 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 multi...