Conflict-driven asp solving with external sources

摘要: Answer Set Programming (ASP) is a well-known problem solving approach based on nonmonotonic logic programs and efficient solvers. To enable access to external information, hex-programs extend with atoms, which allow for bidirectional communication between the program sources of computation (e.g., description reasoners Web resources). Current solvers evaluate by translation ASP itself, in values atoms are guessed verified after ordinary answer set computation. This elegant does not scale number accesses general, particular presence nondeterminism (which instrumental ASP). In this paper, we present novel, native algorithm evaluating uses learning techniques. particular, conflict-driven techniques, prevent solver from running into same conflict again, hex-programs. We show how gain additional knowledge source evaluations use it algorithm. first target uninformed case, i.e., when have no extra information sources, then our case where meta-information available. Experiments that can significantly decrease both runtime considered candidate compatible sets.