To ensure a close relation between the answer sets of a program and those of its ground version, some answer set solvers deal with variables by requiring a safety condition on program rules. If we go beyond the syntax of disjunctive programs, for instance by allowing rules with nested expressions, or perhaps even arbitrary first-order formulas, new definitions of safety are required. In this paper we consider a new concept of safety for formulas in quantified equilibrium logic where answer sets can be defined for arbitrary first-order formulas. The new concept captures and generalises two recently proposed safety definitions: that of Lee, Lifschitz and Palla (2008) as well as that of Bria, Faber and Leone (2008). We study the main metalogical properties of safe formulas. © 2009 Springer Berlin Heidelberg.
CITATION STYLE
Cabalar, P., Pearce, D., & Valverde, A. (2009). A revised concept of safety for general answer set programs. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5753 LNAI, pp. 58–70). https://doi.org/10.1007/978-3-642-04238-6_8
Mendeley helps you to discover research relevant for your work.