PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

A boosting approach to multiple instance learning
Peter Auer and Ronald Ortner
Journal of Machine Learning Research 2005.

This is the latest version of this eprint.


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 metric induced by the infinity-norm these hypotheses can be modified into hyper-rectangles by a greedy algorithm. Compared to other approaches our algorithm delivers improved or at least competitive results on several multiple-instance benchmark data sets.

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Theory & Algorithms
ID Code:1331
Deposited By:Ronald Ortner
Deposited On:28 November 2005

Available Versions of this Item