Abstract
A probabilistic justification of default reasoning is given which is based on ɛ-semantics. Whenever a consequence is derived from a default law and a fact in the presence of further facts, a default assumption is generated claiming that these further facts are probabilistically irrelevant. Each extension Ei of a knowledge base K (in Reiter's sense) corresponds to a set of default assumptions Di generated by Ei. It is proved that if Reiter's original definition is modified, then for each extension Ei, Di will be ɛ-consistent with K, and Ei will contain exactly those sentences which are ɛ-entailed by K plus Di. As a result, we obtain a method of calculating lower probability bounds of the consequences in Ei from the default assumptions Di and the given (nonextreme) lower probability bounds associated with the default laws. Finally, a method of deriving one single and preferred extension E is introduced which is based on a solution of the problem of collective defeat and is probabilistically justified in the same way.
Preview
Unable to display preview. Download preview PDF.
References
E. W. Adams: The Logic of Conditionals. Dordrecht: Reidel 1975.
F. Bacchus et al: Statistical Foundations for Default Reasoning. In: Proc.13th IJCAI 1993, Vol I, pp. 563–569.
G. Brewka: Nonmonotonic Reasoning. Logical Foundations of Commonsense. Cambridge: Cambridge University Press.
R. Carnap: Logical Foundations of Probability. 2nd ed. Chicago: Univ. of Chicago Press 1962.
J.P. Delgrande: An Approach to Default Reasoning Based on a First-Order Conditional Logic: Revised Report. Artificial Intelligence 36, 63–90 (1988).
H. Geffner, J. Pearl: A Framework for Reasoning with Defaults. In: H.E. Kyburg et al (eds.): Knowledge Representation and Defeasible Reasoning. The Netherlands: Kluwer 1990, pp. 69–87.
M. Goldszmidt et al: A Maximum Entropy Approach to Nonmonotonic Reasoning. In: Proc. Nat. Conf. on AI (AAAI-90) 1990, pp. 646–652.
S. Kraus, D. Lehmann and M. Magodor: Nonmonotonic Reasoning, Preferential Models and Cumulative Logics. Artificial Intelligence 44, 167–207 (1990).
D. McDermott/J. Doyle: Non-Monotonic Logic I. Artificial Intelligence 13, 41–72 (1980).
D. Nute: Basic Defeasible Logic. In: L. Farinas del Carro/M. Pentonnen (eds.): Intentional Logics for Programming. Oxford: Oxford University Press 1992, pp. 125–154.
J. Pearl: Probabilistic Reasoning in Intelligent Systems. Santa Mateo: Morgan Kaufmann 1988.
J. L. Pollock: Defeasible Reasoning. Cognitive Science 11, 481–518 (1987).
D. Poole: A Logical Framework for Default Reasoning. Artificial Intelligence 36, 27–47 (1988).
D. Poole: The Effect of Knowledge on Belief: Conditioning, Specifity and the Lottery Paradox in Default Reasoning. Artificial Intelligence 49, 281–307 (1991).
R. Reiter: A Logic for Default Reasoning. Artificial Intelligence 13, 81–132 (1980).
G. Schurz: Erklärungsmodelle in der Wissenschaftstheorie und in der Künstlichen Intelligenz. In: H. Stoyan (ed.): Erklärung im Gespräch — Erklärung im Mensch-Maschine-Dialog.Informatik Fachberichte 310, Berlin: Springer 1992, pp. 1–42.
G. Schurz: Defeasible Reasoning Based on Constructive and Cumulative Rules. In: R. Casati/B. Smith/G. White (eds.): Philosophy and the Cognitive Sciences. Vienna: Hölder-Pichler-Tempsky 1994.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1994 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Schurz, G. (1994). Probabilistic justification of default reasoning. In: Nebel, B., Dreschler-Fischer, L. (eds) KI-94: Advances in Artificial Intelligence. KI 1994. Lecture Notes in Computer Science, vol 861. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-58467-6_22
Download citation
DOI: https://doi.org/10.1007/3-540-58467-6_22
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-58467-4
Online ISBN: 978-3-540-48979-5
eBook Packages: Springer Book Archive