A Framework for Multiple-Instance Learning

TLDR

Multiple‑instance learning is a supervised learning variant where bags of instances are labeled positive if at least one instance belongs to the concept, and negative only if all instances are negative. The authors introduce Diverse Density, a general framework for solving multiple‑instance learning problems. They apply Diverse Density to learn a simple description of a person from image bags, to a stock selection problem, and to drug activity prediction.

Abstract

Multiple-instance learning is a variation on supervised learning, where the task is to learn a concept given positive and negative bags of instances. Each bag may contain many instances, but a bag is labeled positive even if only one of the instances in it falls within the concept. A bag is labeled negative only if all the instances in it are negative. We describe a new general framework, called Diverse Density, for solving multiple-instance learning problems. We apply this framework to learn a simple description of a person from a series of images (bags) containing that person, to a stock selection problem, and to the drug activity prediction problem.