On Dual Programs in Co-Logic Programming
On Dual Programs in Co-Logic Programming
复制标题
DOI：
10.1007/978-3-319-27436-2_2
复制
发表时间：
2015-07
期刊：
影响因子：
--
通讯作者：
H. Seki
中科院分区：
文献类型：
--
作者：
H. Seki
协同逻辑编程是传统逻辑编程语言的扩展，允许每个谓词被注释为归纳或共归纳。为了定义其过程语义以及交替固定点语义，分层限制（程序中谓词依赖性的条件）已强加于协同逻辑程序（co-LP）。在本文中，我们首先考虑协同逻辑编程中的对偶程序：给定一个程序P，它的对偶程序是一个定义P的“补”的程序，即，对于任何基础原子，它计算其否定。当我们考虑具有否定的 co-LP 时，我们表明分层限制通常变得过于严格，并且 Charatonik 等人的 Horn 演算。可以用作协同逻辑编程的扩展，用于处理“非分层”协同LP。然后，我们考虑非分层共同LP在答案集编程（ASP）和有根据的语义（WFS）中的一些应用。特别是，我们通过双程序给出了答案集和 WFS 的新迭代固定点特征。我们还讨论了非分层 co-LP 在程序转换中的一些应用，例如部分演绎和 WFS 的证明程序。
Co-logic programming is an extension of the conventional logic programming language, by allowing each predicate to be annotated as either inductive or coinductive. To define its procedural semantics as well as an alternating fixpoint semantics, thestratification restriction, a condition on predicate dependency in programs, has been imposed on co-logic programs (co-LPs). In this paper, we first considerdualprograms in co-logic programming: Given a programP, its dual programis a program such that it defines the “complement” ofP, i.e., for any ground atom, it computes its negation. When we consider co-LPs with negation, we show that the stratification restriction becomes too restrictive in general, and that the Horn-calculus by Charatonik et al. can be used as an extension of co-logic programming for handling “non-stratified” co-LPs. We then consider some applications of non-stratified co-LPs to Answer Set Programming (ASP) and the well-founded semantics (WFS). In particular, we give new iterated fixpoint characterizations of answer sets as well as the WFS via dual programs. We also discuss some applications of non-stratified co-LPs to program transformation such as partial deduction, and a proof procedure for the WFS.