Multiple Instance Learning by Discriminative Training of Markov Networks
This work addresses the problem of handling ambiguity in weakly supervised data for machine learning practitioners, but it appears incremental as it builds on existing MIL concepts with a new graphical approach.
The authors tackled the problem of multiple instance learning (MIL) by introducing a graphical framework based on Markov networks, which models traditional and general MIL definitions and explores different levels of ambiguity in weakly supervised data, resulting in improved classification performance as verified by experimental results.
We introduce a graphical framework for multiple instance learning (MIL) based on Markov networks. This framework can be used to model the traditional MIL definition as well as more general MIL definitions. Different levels of ambiguity -- the portion of positive instances in a bag -- can be explored in weakly supervised data. To train these models, we propose a discriminative max-margin learning algorithm leveraging efficient inference for cardinality-based cliques. The efficacy of the proposed framework is evaluated on a variety of data sets. Experimental results verify that encoding or learning the degree of ambiguity can improve classification performance.