On finding extensions of default theories
We study from the computational standpoint the problem of finding an extension of a given default theory, and extend previous results in three directions. We show that all default theories that have no odd cycles (in some precise sense) have an extension, which can he found efficiently. We prove that it is NP-complete to find extensions even for default theories with no prerequisites and at most two literals per default. We also characterize precisely the complexity of finding extensions in general default theories: The problem is ∑ 2 P -complete.
Unable to display preview. Download preview PDF.
- [BF]N. Bidoit, C. Froidevaux “General Logical Data-bases and Programs: Default ogic Semantics and Stratification,” Information and Computation, 91, 1, pp. 15–54, 1991.Google Scholar
- [DM]Y. Dimopoulos and V. Magirou “A Graph Theoretic Approach to Default Logic,” to appear in Information and Computation, June 1991.Google Scholar
- [Eth]D. R. Etherington “Formalizing Non-monotonic Reasoning Systems,” Artificial Intelligence, 31, 1987.Google Scholar
- [Go]Georg Gottlob “Complexity Results in Non-monotonic Logics,” CD-TR 91/24, T. U. Wien, August 1991.Google Scholar
- [KP]P. Kolaitis, C. Papadimitriou “Why not Negation by Fixpoint?” Proc. 1988 PODS. Also, JCSS 1991.Google Scholar
- [KS]H. Kautz and B. Selman “Hard Problems for Simple Default Logics,” Artificial Intelligence, 1991.Google Scholar
- [PY]C. H. Papadimitriou and M. Yannakakis “Tiebreaking Semantics and Structural Totality of Logic Programs with Negation,” Proceedings of 11th PODS Conference, San Diego, June 1992.Google Scholar
- [Re]R. Reiter “A Logic for Default Reasoning,” Artificial Intelligence 13, 1980.Google Scholar
- [Se]B. Selman “The Complexity of Non-monotonic Reasoning” Ph.D. Thesis, Toronto Univ., 1990.Google Scholar