paperbot · PL 论文追踪

全部论文 待读 收藏

P2 Decidability and Complexity in Weakening and Contraction Hypersequent Substructural Logics

弱化与收缩超序列子结构逻辑中的可判定性与复杂度
LICS 2021 36th Annual ACM/IEEE Symposium on Logic in Computer Science (LICS)2021匹配度 1/10 · 该论文研究子结构逻辑的超序列演算可判定性与复杂度,属于纯逻辑/证明论方向,未涉及程序静态分析、LLM/Agent、浮点程序分析或贝叶斯程序分析,与研究画像核心问题基本无关。引用 3 inbox
A. R. Balasubramanian, Timo Lang, Revantha Ramanayake
crossref: 2026-09-12

AI 速览

一句话:证明了具有无割超序列演算的所有 HFLew 解析结构规则扩张可判定,并给出弱化与收缩扩张的复杂度上界及 MTL 的首个复杂度界。

本文研究交换 Full Lambek 逻辑带弱化 FLew(即 IMALLW)的无穷多公理扩张的可判定性与复杂度。作者证明:HFLew 的每个解析结构规则扩张都可判定,从而覆盖所有具有无割超序列证明演算的扩张;其收缩对应物 FLec 的相应扩张虽已知可判定,但复杂度此前未知。为此,文章引入面向良拟序的长度函数与快速增长复杂度类工具,为弱化和收缩两类扩张给出复杂度上界。作为一个具体实例,该结果首次给出著名模糊逻辑 MTL 的复杂度界,解决了长期开放问题。

问题
研究带弱化的交换 Full Lambek 逻辑 FLew(IMALLW)及其收缩对应物 FLec 的无穷多公理扩张中,哪些可判定,以及已有可判定性结果的扩张其计算复杂度如何。
方法
利用无割超序列演算和解析结构规则,结合良拟序的长度函数与快速增长复杂度类,分析弱化与收缩扩张。
评估
通过证明 HFLew 所有解析结构规则扩张可判定,并给出弱化与收缩扩张的复杂度上界,尤其首次确定模糊逻辑 MTL 的复杂度界。
原文摘要(Abstract)

We establish decidability for the infinitely many axiomatic extensions of the commutative Full Lambek logic with weakening FLew (i.e. IMALLW) that have a cut-free hypersequent proof calculus. Specifically: every analytic structural rule extension of HFLew. Decidability for the corresponding extensions of its contraction counterpart FLec was established recently but their computational complexity was left unanswered. In the second part of this paper, we introduce just enough on length functions for well-quasi-orderings and the fast-growing complexity classes to obtain complexity upper bounds for both the weakening and contraction extensions. A specific instance of this result yields the first complexity bound for the prominent fuzzy logic MTL (monoidal t-norm based logic) providing an answer to a longstanding open problem.

链接与引用

DOI 原文 · PDF(开放获取) ·

BibTeX
@inproceedings{paperbot6194,
  title = {Decidability and Complexity in Weakening and Contraction Hypersequent Substructural Logics},
  author = {A. R. Balasubramanian and Timo Lang and Revantha Ramanayake},
  booktitle = {LICS},
  year = {2021},
  doi = {10.1109/lics52264.2021.9470733}
}