Circumscriptive Event Calculus as Answer Set Programming

Recently, Ferraris, Lee and Lifschitz presented a general definition of a stable model that is similar to the definition of circumscription, and can even be characterized in terms of circumscription. In this paper, we show the opposite direction, which is, how to turn circumscription into the general stable model semantics, and based on this, how to turn circumscriptive event calculus into answer set programs. The reformulation of the event calculus in answer set programming allows answer set solvers to be applied to event calculus reasoning, handling more expressive reasoning tasks than the current SAT-based approach. Our experiments also show clear computational advantages of the answer set programming approach.