Dempster-Shafer Theory (DST) generalizes Bayesian probability theory, but suffers from a high computational burden . A lot of work has been done to reduce the complexity of computationsused in information fusion with Dempsters’s rule . The main approaches exploiteither the structure of Boolean lattices or the information contained in beliefsources . In this paper, we propose sequences of graphs for the computation of the zeta and M\”obiustransformations that optimally exploit both the structure . of distributivesemilattices and the information . contained in belief sources . We call them theEfficient M\”obius Transformations (EMT) The complexity of theEMT is always inferior to the complexity . of algorithms that consider the wholelattice, such as the Fast FMT for all DST transformations . More generally, ourEMTs apply to any function in any finite distributive

Author(s) : Maxime Chaveroche, Franck Davoine, Véronique Cherfaoui

Links : PDF - Abstract

Code :
Coursera

Keywords : information - theory - complexity - transformations - contained -

Leave a Reply

Your email address will not be published. Required fields are marked *