Un point de vue épistémique sur la fusion de croyances : l’identification du monde réel

Belief merging is often described as the process of defining a belief base which best represents the beliefs of a group of agents (a profile of belief bases). The resulting belief base can be viewed as a synthesis of the input profile. In this paper another view, called the epistemic view, of what belief merging aims at is considered. Under this view the purpose of belief merging is to best approximate what is the true state of the world. We study generalizations of Condorcet’s Jury Theorem from the belief merging perspective. Roughly we show that if we merge the beliefs of sufficiently many reliable agents then we can ensure to identify the true state of the world. We show that some (but not all) merging operators from the literature are suited to the truth tracking issue.