有限复杂度饱和序的自动综合
Automated Synthesis of a Finite Complexity Ordering for Saturation
摘要:
本文提出了一种新的方法来饱和一组子句,使得A < B对每个原子A和B都蕴涵Var(A){ \subseteq } Var(B).这个条件是由任何原子排序与字典序,递归,或多集路径排序项兼容。我们的饱和过程是基于先验有序分辨率和其主要的新颖性是在飞行中建设的有限复杂性原子排序。与通常的冗余相反,我们给出了一个新的冗余概念,并证明了在饱和期间,先验有序分辨率的非冗余推理也是后验有序分辨率的推理。我们还证明了,如果一组S的条款是饱和的原子排序如上所述,那么一个条款C是否从S是可判定的问题。
英文摘要:
We present in this paper a new procedure to saturate a set of clauses with respect to a well-founded ordering on ground atoms such that A < B implies Var(A) {\subseteq} Var(B) for every atoms A and B. This condition is satisfied by any atom ordering compatible with a lexicographic, recursive, or multiset path ordering on terms. Our saturation procedure is based on a priori ordered resolution and its main novelty is the on-the-fly construction of a finite complexity atom ordering. In contrast with the usual redundancy, we give a new redundancy notion and we prove that during the saturation a non-redundant inference by a priori ordered resolution is also an inference by a posteriori ordered resolution. We also prove that if a set S of clauses is saturated with respect to an atom ordering as described above then the problem of whether a clause C is entailed from S is decidable.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。