PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

A boosting approach to multiple instance learning
Peter Auer and Ronald Ortner
In: ECML 2004, 20-24 Sep 2004, Pisa, Italy.

There is a more recent version of this eprint available. Click here to view it.


In this paper we present a boosting approach to multiple instance learning. As weak hypotheses we use balls (with respect to various metrics) centered at instances of positive bags. For the infinity-norm these hypotheses can be modified into hyper-rectangles by a greedy algorithm. Our approach includes a stopping criterion for the algorithm based on estimates for the generalization error. These estimates can also be used to choose a preferable metric and data normalization. Compared to other approaches our algorithm delivers improved or at least competitive results on several multiple instance benchmark data sets.

EPrint Type:Conference or Workshop Item (Paper)
Additional Information:Available also from
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Theory & Algorithms
ID Code:537
Deposited By:Ronald Ortner
Deposited On:25 December 2004

Available Versions of this Item