Computing Intersection of Autoepistemic Expansions.

摘要: In this paper, we consider the question of skeptical reasoning for an important nonmonotonic system — autoepistemic logic Moore. Autoepistemic is a method which assigns to set formulas collection theories called stable expansions. A naive perform deciding whether given formula φ belongs all expansions theory compute first and then check each them. This approach however prohibitively inefficient. The goal paper propose different computing intersection theory. Our does not require us any expansion It reduces membership in propositional provability. More precisely, describe that modal I PI modal-free another ′ such manner if only ⊢ . general, much larger than original I. We have found, however, several cases when it so size polynomial These classes are closely related programs disjunctive programs. Consequently, obtain methods atom supported (or stable) models (disjunctive) program, as well numerous complexity results.