A series of revisions of David Poole’s specificity
A series of revisions of David Poole’s specificity
复制标题
大卫·普尔的特殊性的一系列修订
DOI：
--
复制
发表时间：
2016
影响因子：
1.2
通讯作者：
Frieder Stolzenburg
中科院分区：
文献类型：
--
作者：
Claus;Frieder Stolzenburg
在20世纪80年代中期，大卫普尔（David Poole）向人工智能界引入了一个语义的、模型理论的特异性概念。从那时起，它在非单调推理中得到了进一步的应用，特别是在可废止推理中。普尔试图接近人类直觉的特异性概念，这似乎是必不可少的推理在日常生活中与其部分和不一致的信息。然而，他的概念被证明是复杂和有问题的，正如我们所展示的，在某种程度上可以通过更接近人类直觉的特异性概念来克服。除了直观的优势，我们的新的特异性订购普尔的特异性关系在经典的例子中的文献，我们还报告了一些硬的数学事实：相反，什么是声称之前，我们表明，普尔的关系是不传递的一般。我们的第一个特异性排序（CP1）在纠正了普尔的技术缺陷之后，尽可能接近地捕捉到了它最初的直觉。第二个（CP2）是CP1的变体，并提出了一个类似的概念，最终可能解决普尔式的特异性关系的棘手问题的一步。目前的手段，决定我们的新的特异性关系，但是，显示只有轻微的改进，比已知的普尔的关系，因此，我们建议在实践中的应用程序更有效的变通办法。
In the middle of the 1980s, David Poole introduced a semantic, model-theoretic notion of specificity to the artificial-intelligence community. Since then it has found further applications in non-monotonic reasoning, in particular in defeasible reasoning. Poole tried to approximate the intuitive human concept of specificity, which seems to be essential for reasoning in everyday life with its partial and inconsistent information. His notion, however, turns out to be intricate and problematic, which — as we show — can be overcome to some extent by a closer approximation of the intuitive human concept of specificity. Besides the intuitive advantages of our novel specificity orderings over Poole’s specificity relation in the classical examples of the literature, we also report some hard mathematical facts: Contrary to what was claimed before, we show that Poole’s relation is not transitive in general. The first of our specificity orderings (CP1) captures Poole’s original intuition as close as we could get after the correction of its technical flaws. The second one (CP2) is a variation of CP1 and presents a step toward similar notions that may eventually solve the intractability problem of Poole-style specificity relations. The present means toward deciding our novel specificity relations, however, show only slight improvements over the known ones for Poole’s relation; therefore, we suggest a more efficient workaround for applications in practice.
RatioLog 项目：逻辑推理的理性扩展
DOI：
10.1007/s13218-015-0377-9
发表时间：
2015
期刊：
KI - Künstliche Intelligenz
影响因子：
--
作者：
Ulrich Furbach;Claudia Schon;Frieder Stolzenburg;Karl-Heinz Weis;Claus-Peter Wirth
通讯作者：
Claus-Peter Wirth