Set-valued Bayesian Inference with Probabilistic Equivalence

In this paper, a unified view of the problem of class-selection with Bayesian classifiers is presented. Select-ing a subset of classes instead of singleton allows 1) to reduce the error rate and 2) to propose a reduced set to another classifier or an expert. This second step pro-vides additional information, and therefore increases the quality of the result. The proposed framework, based on the evaluation of the probabilistic equiva-lence, allows to retrieve the class-selective frameworks that have been proposed in the literature. Several ex-periments show the effectiveness of this generic propo-sition. 1. Introduction an