Belief Base Rationalization for Propositional Merging
Sebastien Konieczny, Pierre Marquis and Nicolas Schwind
Belief merging operators aim at defining a base which represent the beliefs of a group of agents given their individual belief bases. Integrity constraints, representing physical laws or norms, are often used in the merging process. Existing operators take advantage of all the pieces of information from the bases, including those contradicting the integrity constraints, which is not suited to every merging scenario. In this paper, we study the case where the bases are "rationalized" during the merging process with respect to the integrity constraints. We define in formal terms several independence conditions for merging operators with respect to different forms of rationalization and study how they interact with the standard (IC) postulates for belief merging. Especially, we give an independence-based axiomatic characterization for the distance-based operator $\Delta_{d_D, \Sigma}$.