102 search results - page 1 / 21 » A Faster Algorithm for Generalized Multiple-Instance Learnin... |
FLAIRS
2004
13 years 5 months ago
2004
In our prior work, we introduced a generalization of the multiple-instance learning (MIL) model in which a bag's label is not based on a single instance's proximity to a...
ICTAI
2006
IEEE
13 years 10 months ago
2006
IEEE
We present MI-Winnow, a new multiple-instance learning (MIL) algorithm that provides a new technique to convert MIL data into standard supervised data. In MIL each example is a co...
ECML
2004
Springer
13 years 9 months ago
2004
Springer
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...
ECCV
2010
Springer
13 years 4 months ago
2010
Springer
Abstract. Multiple-instance learning (MIL) allows for training classifiers from ambiguously labeled data. In computer vision, this learning paradigm has been recently used in many ...
NIPS
2003
13 years 5 months ago
2003
Learning from ambiguous training data is highly relevant in many applications. We present a new learning algorithm for classification problems where labels are associated with se...