Applying the Local Computation Framework to Classical and Non-Classical Logics
Jérôme Mengin, Nic Wilson · 1997
Algorithms to propagate uncertainty using Shenoy and Shafer's Local Computation framework give rise to efficient computation in a number of spheres of reasoning, such as Bayesian probability, DempsterShafer Belief, infinitesimal probability functions, and Zadeh's Possibility functions. Finite sets of possibilities (or constraints) can be propagated with this framework, and so deduction in a finite propositional calculus can be performed by considering sets of possible worlds. Recently Kohlas and Moral have shown how to use Local Computation to directly propagate sets of formulae in clausal form in a finite propositional calculus. In this report we explore the application of the Local Computation framework to other logics. We describe how a logic can be embedded in the framework, given that its semantics verifies certain properties. This is applied to first-order predicate calculus, modal and conditional logics. We also discuss its application to predicate circumscription....