Conflict-driven non-ground rule learning

[rtaupe]

Alpha'a predecessor OMiGA employs unfolding and shifting techniques to learn new non-ground rules during the solving process, thereby improving search performance [1]. Alpha would probably also benefit from such techniques (at least from unfolding). Rules learnt that way can also be added to the input program to improve future solver runs.

[1]: Weinzierl, Antonius (2013): Learning Non-Ground Rules for Answer-Set Solving. In David Pearce, Shahab Tasharrofi, Evgenia Ternovska, Concepción Vidal (Eds.): 2nd Workshop on Grounding and Transformations for Theories With Variables. Workshop on Grounding and Transformations for Theories With Variables. Corunna, Spain, September 15, pp. 25–37.