Belief change in branching time: AGM-consistency and iterated revision (Q421352): Difference between revisions

The author continues his study of belief change in the context of branching-time models, begun in the paper [``Axiomatic characterization of the AGM theory of belief revision in a temporal logic'', Artif. Intell. 171, No. 2--3, 144--160 (2007; Zbl 1168.03317)]. There, the AGM postulates for belief revision were represented in a language containing three modal operators: a belief operator, one representing the exact amount of information available to an agent and, crucially, a `next-time' operator in branching time. The language thus differs from those of the dynamic doxastic logics that have also been used to represent belief change in a modal setting. The paper under review studies the conditions under which a partial belief revision function induced by a state-time pair in a frame may be extended to a belief revision function defined over the entire frame and satisfying the AGM postulates. Final sections also explore the formulation of well-known principles of iterated belief change using the temporally modalized language.