Decidability of Contextual Reasoning

Vanja Buvac

Contexts were first suggested in McCarthy’s Turing Award Paper, (McCarthy 1987), as a possible solution to the problem of generality in AI. McCarthy’s concern with the existing AI systems has been that they can reason only about some particular, predetermined task. When faced with slightly different circumstances they need to be completely rewritten. In other words, AI systems lack generality. Cyc (Guha and Lenat 1990), a large common-sense knowledge-base currently being developed at MCC, is one example of where contexts have already been put to use in attempt to solve the problem of generality. Because of the complexity of the problem of generality, it has been speculated that any reasoning system which would be able to solve this problem would itself be computationally unacceptable. The purpose of this paper is to show that propositional contextual reasoning is decidable.

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.