New inference rules for efficient Max-SAT solving
Federico Heras, Javier Larrosa · 2006
In this paper we augment the Max-SAT solver of (Larrosa & Heras 2005) with three new inference rules. The three of them are special cases of Max-SAT resolution with which bet-ter lower bounds and more value pruning is achieved. Our ex-perimental results on several domains show that the resulting algorithm can be orders of magnitude faster than state-of-the-art Max-SAT solvers and the best Weighted CSP solver.