MIML: A Framework for Learning with Ambiguous Objects

In this paper, we propose the MIML (Multi-Instance Multi-Label learning) framework for learning with ambiguous objects, where an example is described by multiple instances and associated with multiple class labels. Comparing with traditional learning frameworks, the MIML framework is more convenient and natural for representing ambiguous objects. To learn MIML examples, we propose the MimlBoost and MimlSvm algorithms based on a simple degeneration strategy, and experiments show that solving problems involving ambiguous objects in the MIML framework can lead to good performances. Considering that the degeneration process may lose information, we propose the D-MimlSvm algorithm which tackles MIML problems directly in a regularization framework. Moreover, we show that even when we do not have access to the raw objects and thus cannot capture more information from raw objects by using the MIML representation, MIML is still useful. We propose the InsDif and SubCod algorithms. InsDif works by transforming single-instances into the MIML representation for learning, while SubCod works by transforming single-label examples into the MIML representation for learning. Experiments show that in some tasks they are able to achieve better performances than learning the single-instances or single-label examples directly.