2011/09/29 by Frédéric Dambreville, Frederic Dambreville, Dambreville, Frederic
Computer Science · Decision Sciences · Mathematics · #Artificial Intelligence (cs.AI) #Bayesian Modeling and Causal Inference #FOS: Computer and information sciences #FOS: Mathematics #Logic (math.LO) #Logic in Computer Science (cs.LO) #Logic, Reasoning, and Knowledge #Multi-Criteria Decision Making #cs.AI #cs.LO #math.LO
paper · pdf · doi:10.48550/arxiv.1109.6401
arxiv created 2011/09/29 · openalex publication_date 2011/09/29 · arxiv updated 2011/10/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
While belief functions may be seen formally as a generalization of probabilistic distributions, the question of the interactions between belief functions and probability is still an issue in practice. This question is difficult, since the contexts of use of these theory are notably different and the semantics behind these theories are not exactly the same. A prominent issue is increasingly regarded by the community, that is the management of the conflicting information. Recent works have introduced new rules for handling the conflict redistribution while combining belief functions. The notion of conflict, or its cancellation by an hypothesis of open world, seems by itself to prevent a direct interpretation of belief function in a probabilistic framework. This paper addresses the question of a probabilistic interpretation of belief functions. It first introduces and implements a theoretically grounded rule, which is in essence an adaptive conjunctive rule. It is shown, how this rule is derived from a logical interpretation of the belief functions by means of a probabilistic multimodal logic; in addition, a concept of source independence is introduced, based on a principle of entropy maximization.