UniLog: A Unified Logic Framework for Constraint Representation and Handling in Intelligent Systems

Read full text Summarise with VIRAT
Authors: Harris Wang
Cited by: 0
References: 28

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This paper presents UniLog, a unified logic framework designed to integrate the expressive power and reasoning capabilities of diverse logical systems into a single coherent formalism. UniLog is specifically developed to serve as the constraint representation language for the Constrained Object Hierarchies (COH) and Constrained Zone-Object Architecture (CZOA) frameworks, enabling the specification of identity constraints, trigger constraints, goal constraints, and access constraints in a unified manner. The core novelty of UniLog lies not in any single logic but in its architecture: a modular, fibred system that combines a core modal‑temporal substructure with conservatively extended modules for probabilistic, non‑monotonic, epistemic, and domain‑specific reasoning. We provide formal syntax, semantics, and inference rules for UniLog, demonstrate its embedding of major logic families, and illustrate its application to complex organizational intelligent information systems. In addition, we establish key meta‑theoretic properties: conservativity of modular extensions (Theorem 3), soundness of the inference system, decidability of useful fragments, and soundness and relative completeness of a translation from a bounded fragment of UniLog to SMT‑LIB. These results provide a rigorous foundation for using UniLog in formal verification and for integrating it with automated theorem provers.

Fields of Study & Subjects

Logic Reasoning and Knowledge programming and type systems Constraint Satisfaction and Optimization

Publication details

Journal / source
Preprints.org
Publication date
26 March 2026
Language
EN
Document type
Preprint