Generalizing Modular Logic Programs
This work addresses a practical limitation in modular Answer Set Programming for researchers and practitioners, representing an incremental improvement to an existing framework.
The paper tackles the restrictive module conditions in Oikarinnen and Janhunen's Gaifman-Shapiro-style architecture for Answer Set Programming, which limit practical applicability by requiring disjoint output signatures and forbidding positive cyclic dependencies. It proposes alternative ways to lift both restrictions independently, solving the first and widening the framework's applicability and the scope of the module theorem.
Even though modularity has been studied extensively in conventional logic programming, there are few approaches on how to incorporate modularity into Answer Set Programming, a prominent rule-based declarative programming paradigm. A major approach is Oikarinnen and Janhunen's Gaifman-Shapiro-style architecture of program modules, which provides the composition of program modules. Their module theorem properly strengthens Lifschitz and Turner's splitting set theorem for normal logic programs. However, this approach is limited by module conditions that are imposed in order to ensure the compatibility of their module system with the stable model semantics, namely forcing output signatures of composing modules to be disjoint and disallowing positive cyclic dependencies between different modules. These conditions turn out to be too restrictive in practice and in this paper we discuss alternative ways of lift both restrictions independently, effectively solving the first, widening the applicability of this framework and the scope of the module theorem.