Integration of evolutionary prejudices in Dempster-Shafer theory
This paper does two things:\N\begin{itemize}\N\item[1.] It distinguishes between prejudices and evidence in the context of belief change claiming that prejudices influence the evolution of beliefs at the meta-level and evidence influences the evolution of beliefs at the object-level.\N\item[2.] Within the framework of Dempster-Shafer theory, it proposes a formal description of belief change that obeys its distinction between the roles of prejudices and evidence.\N\end{itemize}\N\NThe authors define a prejudice as an adopted opinion that acts as an a priori weight that favors or resists certain beliefs. Besides presenting a formal theory, the authors do not explain in detail what they mean by their distinction between meta-level and object-level (a term that they do not use but that I take to be the opposite of meta-level). That prejudices operate at the meta-level seems to mean that an agent would not say ``I believe in proposition \(P\)`` (where \(P\) is a prejudice) but they would rather say ``\(P\) predisposes me to weight (favor or resist) certain updates to my beliefs. Conversely, that evidence operates at the object-level seems to mean that a new piece of evidence ``directly allocates support to propositions. Allocation of support is governed by a mass function over the subsets of the set of descriptions of possible situations. Prejudices shape the evolution of beliefs acting as biases rather than modification of the mass function itself.\N\NWithin this framework, one separates what one believes (governed by the mass function) from how one is predisposed to update what one believes (the prejudices). On this basis, the authors model prejudices of different types: persistent (prejudices that never evolve), strong (prejudices that forbid changing some beliefs), malleable (e.g., prejudices that slowly decrease when confronted with contradictory evidence), etc.\N\NAt some point, the authors muddle the distinction between meta-level and object-level by saying ``commensurability between evidence and prejudice strengths is imposed[,] losing a little of the meta character of the phenomenon (page 4). They do not explain this point further.
- A logic for default reasoning
- Conditional independence in valuation-based systems
- Conjunctive and disjunctive combination of belief functions induced by nondistinct bodies of evidence
- Defining belief functions using mathematical morphology - application to image fusion under imprecision
- Face pixel detection using evidential calibration and fusion
- scientific article; zbMATH DE number 6093230 (Why is no real title available?)
- scientific article; zbMATH DE number 3560492 (Why is no real title available?)
- scientific article; zbMATH DE number 1269589 (Why is no real title available?)
- scientific article; zbMATH DE number 3271180 (Why is no real title available?)
- Measure-theoretic approach to the inversion problem for belief functions
- Nonmonotonic reasoning, conditional objects and possibility theory
- On Latent Belief Structures
- On the logic of theory change: Partial meet contraction and revision functions
- On the unicity of dempster rule of combination
- Prejudice in uncertain information merging: pushing the fusion paradigm of evidence theory further
- Propositional knowledge base revision and minimal change
- Qualitative capacities: basic notions and potential applications
- Qualitative probabilities for default reasoning, belief revision, and causal modeling
This page was built for publication: Integration of evolutionary prejudices in Dempster-Shafer theory
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7016605)
