We describe the transferable belief model, a model for representing quantified beliefs based on belief functions. Beliefs can be held at two levels: (1) a credal level where beliefs are entertained and quantified by belief functions, (2) a pignistic level where beliefs can be used to make decisions

Markov random field models provide a robust and unified framework for early vision problems such as stereo, optical flow and image restoration. Inference algorithms based on graph cuts and belief propagation yield accurate results, but despite recent advances are often still too slow for practical

for cognitively more appropriate operations. Furthermore, it makes implementation more feasible. In this paper we show how extra structure of belief bases can be used for implementing local change as defined in [ Hansson and Wassermann, 1999 ] , where only the relevant part of an agent's beliefs

Important inference problems in statistical physics, computer vision, error-correcting coding theory, and artificial intelligence can all be reformulated as the computation of marginal probabilities on factor graphs. The belief propagation (BP) algorithm is an efficient way to solve these problems

The Recovery postulate for contraction says that any beliefs lost due to the contraction of some belief p should return if p is immediately re-asserted. Recovery holds for logically closed sets of beliefs, but it does not hold for belief bases (sets of beliefs that are not logically closed

Abstract Recently, researchers have demonstrated that "loopy belief propagation" -the use of Pearl's polytree algorithm in a Bayesian network with loops -can perform well in the context of error-correcting codes. The most dramatic instance of this is the near Shannon

We look at the problem of revising fuzzy belief bases, i.e., belief base revision in which both formulas in the base as well as revisioninput formulas can come attached with varying truth-degrees. Working within a very general framework for fuzzy logic which is able to capture certain types

Many coherence-based approaches to inconsistency handling within propositional belief bases have been proposed so far. They consist in selecting one or several preferred consistent subbases of the given (usually inconsistent) stratified belief base (SBB), then using classical inference from some

for belief bases by means of a construction and a set of postulates and show a representation result. We also illustrate the use of this operation to formalize abductive diagnosis. 1

In this paper, we investigate the extent to which knowledge compilation can be used to circumvent the complexity of skeptical inference from a stratified belief base (SBB). We first analyze the compilability of skeptical inference from an SBB, under various requirements concerning both