Stephen D. Scott, Jun Zhang, and Joshua Brown. On Generalized Multiple-Instance Learning. Technical report UNL-CSE-2003-5, University of Nebraska, 2003.
Abstract
449Kb PDF

Abstract

We describe a generalization of the multiple-instance learning model in which a bag's label is not based on a single instance's proximity to a single target point. Rather, a bag is positive if and only if it contains a collection of instances, each near one of a set of target points. We list potential applications of this model (robot vision, content-based image retrieval, protein sequence identification, and drug discovery) and describe target concepts for these applications that cannot be represented in the conventional multiple-instance learning model. We then adapt a learning-theoretic algorithm for learning in this model and present empirical results.

Keywords: multiple-instance learning, Hausdorff metric, Winnow, content-based image retrieval, drug discovery, protein sequence identification

To Stephen D. Scott's home page

Last modified 30 May 2003.