Computational Aspects of the Calculus of Structure
Universidade Federal de Minas Gerais(米纳斯吉拉斯联邦大学)
Mário S. Alvim
备注:
Thesis presented by Mário S. Alvim as part of the requirements for the degree or Master of Science in Computer Science granted by the Universidade Federal de Minas Gerais. April, 04th, 2008
逻辑是正确推理的科学,逻辑系统是用正确的方法证明一定逻辑中的断言的工具。有许多逻辑系统,以及许多形式化它们的方法,例如,使用自然演绎或微积分。结构演算(Calculus of Structures,CoS)是由Alessio Guglielmi在2004年提出的一种新的形式主义,它在推理规则可以应用于公式内部的任何深度的意义上推广了结构演算,而不仅仅是主要的连接词。有了这个特性,CoS中的证明比任何其他支持分析证明的形式主义都要短。虽然拥有CoS的自由和表现力是很好的,但从证明搜索的角度来看,更多的自由意味着更大的搜索空间。在寻找演绎系统的完全自动化时,这应该受到限制。人们做了一些努力来减少这种非确定性,但它们基本上都是可操作的方法,到目前为止,还没有关于CoS计算行为的坚实的理论结果。本论文的主要焦点是讨论如何提出一个适合于实现的CoS证明搜索策略。这一战略应该是理论性的,而不是纯粹的操作性的。我们引入了结构内部子结构的不相干数的概念,并使用这个概念来实现我们的主要结果:根据我们的猜想,有一个算法对应于一个证明搜索策略,每个可证明结构的子系统中的FBV(乘法线性逻辑MLL加上规则混合)只包含成对不同的原子。我们的算法实现,我们相信我们的策略是一个很好的起点,利用计算方面的CoS在更一般的系统,如BV本身。
英文摘要:
Logic is the science of correct inferences and a logical system is a tool to prove assertions in a certain logic in a correct way. There are many logical systems, and many ways of formalizing them, e.g., using natural deduction or sequent calculus. Calculus of structures (CoS) is a new formalism proposed by Alessio Guglielmi in 2004 that generalizes sequent calculus in the sense that inference rules can be applied at any depth inside a formula, rather than only to the main connective. With this feature, proofs in CoS are shorter than in any other formalism supporting analytical proofs. Although it is great to have the freedom and expressiveness of CoS, under the point of view of proof search more freedom means a larger search space. And that should be restricted when looking for complete automation of deductive systems. Some efforts were made to reduce this non-determinism, but they are all basically operational approaches, and no solid theoretical result regarding the computational behaviour of CoS has been achieved so far. The main focus of this thesis is to discuss ways to propose a proof search strategy for CoS suitable to implementation. This strategy should be theoretical instead of purely operational. We introduce the concept of incoherence number of substructures inside structures and we use this concept to achieve our main result: there is an algorithm that, according to our conjecture, corresponds to a proof search strategy to every provable structure in the subsystem of FBV (the multiplicative linear logic MLL plus the rule mix) containing only pairwise distinct atoms. Our algorithm is implemented and we believe our strategy is a good starting point to exploit the computational aspects of CoS in more general systems, like BV itself.