Nonmonotonic causal logic, invented by McCain and Turner, is a formalism well suited for representing knowledge about actions, and the definite fragment of that formalism has been implemented in the reasoning and planning system called CCalc. A 1997 theorem due to McCain shows how to translate definite causal theories into logic programming under the answer set semantics, and thus opens the possibility of using answer set programming for the implementation of such theories. In this paper we propose a generalization of McCain’s theorem that extends it in two directions. First, it is applicable to arbitrary causal theories, not only definite. Second, it covers causal theories of a more general kind, which can describe non-Boolean fluents.

PDF

In IJCAI 2007.

@inproceedings{ferp07, title={A Logic Program Characterization of Causal Theories}, author={Paolo Ferraris}, booktitle={IJCAI}, url="http://www.cs.utexas.edu/users/ai-lab?ferp07", year={2007} }