首页> 外文会议>International Workshop on Logic, Language, Information and Computation >On Infinitary Proof Theory of Logics of Information and Common Belief
【24h】

On Infinitary Proof Theory of Logics of Information and Common Belief

机译:信息逻辑与共同信念的无限证明论

获取原文

摘要

Recently there has been a growing interest in applying non-classically based modal logics in the context of logics for agency and social behaviour. In particular, substructural or other information-based modal logics of knowledge and belief, or similar versions of PDL, have been designed. While basic modal extensions of substructural logics on one side, and classically based logics of common belief and other fixed point modalities, are relatively well understood when it comes to complete6ness and proof theory, with logics we have in mind it is not so. In this talk, we will mainly concentrate on logics of common belief. We will consider two natural ways of axiomatizing the common belief over a basic modal logic (Belnap-Dunn logic or distributive substructural logics, extended by normal diamond and box modalities): one finitary, which is the standard Kozen's axiomatization, and the other infinitary, with an infinitary rule replacing the induction rule and using finite approximations of the fixed points. The finitary axiomatization is used to obtain, using an algebraic (and coalgebraic) insight, the soundness of the infinitary rule. We will then concentrate on the infinitary part of the story and draw a general, duality based picture connecting the syntax and poset-based frame semantics of the infinitary axiomatizations, including a completeness argument based on a canonical model construction. Here, the infintary case differs from the usual finitary account of (non-classical) modal logics: in particular, one needs to use an appropriate version of Lindenbaum or Belnap Pair-Extension lemma. Finally, we use the above insight to discuss proof theory of such logics within the framework of display calculi.
机译:最近,人们越来越关注在代理和社会行为逻辑的背景下应用基于非经典的模态逻辑。特别是,已经设计了知识和信念的子结构或其他基于信息的模态逻辑,或类似版本的PDL。一方面,子结构逻辑的基本模态扩展,以及基于共同信念和其他不动点模态的经典逻辑,在完整性和证明理论方面相对较容易理解,但我们认为逻辑并非如此。在本演讲中,我们将主要集中于共同信念的逻辑。我们将考虑两种对基本模态逻辑(贝纳普-邓恩逻辑或分布子结构逻辑,由正常的菱形和盒子模态扩展的公有信念)的公理化的自然方法:一种是最终的,这是标准的Kozen的公理化,另一种是不定式的,用不定式规则代替归纳规则并使用不动点的有限近似。最终公理化用于使用代数(和cogegebraic)见解来获得不定律的合理性。然后,我们将集中于故事的非限定性部分,并绘制一个基于对偶性的一般图片,以连接非限定性公理化的语法和基于姿态的框架语义,包括基于规范模型构造的完整性论证。在这里,无数情况不同于(非经典)模态逻辑的通常最终解释:特别是,人们需要使用Lindenbaum或Belnap Pair-Extension引理的适当版本。最后,我们利用以上见解在显示计算框架内讨论此类逻辑的证明理论。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号