AGL: actionable granular logic for verifiable specifications and reasoningin AI decision–action systems
Andrzej Jankowski
a:1:{s:5:"en_US";s:3:"UWM";}Abstrakt
Decision–action systems deployed in high-stakes domains require auditability and bounded (profiled) verifiabil-ity of the reasoning core, which cannot be ensured by empirical safeguards alone. We introduce AGL (Actionable Granular Logic) as a formal framework that combines an auditable knowledge-state layer (MT-FOGL) with work-flows described by a regular-program syntax in the style of FO-PDL. Vague and probabilistic assessments are encapsulated as Information Granules and exposed to the core solely via threshold atoms, thereby keeping the rule/procedure interface within classical two-valued logic. To control verifiability, we restrict quantification and program tests to guarded profiles GF/RGF, preserving decidability with known worst-case complexity bounds. Complex estimation mechanisms are deliberately kept outside the verifiable core (the Decidability Split). The approach is illustrated using non-normative decision patterns in a medical context, intentionally independent of any particular clinical guideline version.
Słowa kluczowe:
Actionable Granular Logic, LLM Hallucinations, Guarded Fragments, Neuro-symbolic AI, Clinical Decision Support, Verifiable AI, OnkoBotBibliografia
Andréka, H., Németi, I., van Benthem, J. 1998. Modal Languages and Bounded Fragments of Predicate Logic.
Crossref
Google Scholar
Journal of Philosophical Logic, 27(3), 217–274. Google Scholar
Bargiela, A., Pedrycz, W. 2003. Granular Computing: An Introduction. Kluwer Academic Publishers. Bednarczyk, B., Kieroński, E. 2025. Guarded Fragments Meet Dynamic Logic: The Story of Regular Guards. In Google Scholar
Proceedings of the 22nd International Conference on Principles of Knowledge Representation and Reasoning (KR 2025), 89–99. Google Scholar
Briganti, A., et al. 2019. A novel tool to predict lymph node metastases in prostate cancer: the 2018 Briganti nomogram. European Urology Oncology, 2(4), 420–426. Google Scholar
Grädel, E. 1999. On the Restraining Power of Guards. The Journal of Symbolic Logic, 64(4), 1719–1742. Harel, D., Kozen, D., Tiuryn, J. 2000. Dynamic Logic. MIT Press.
Crossref
Google Scholar
Jankowski, A. 2017. Interactive Granular Computations in Networks and Systems Engineering: A Practical Perspective. Springer.
Crossref
Google Scholar
Pawlak, Z. 1982. Rough Sets. International Journal of Computer & Information Sciences, 11, 341–356.
Crossref
Google Scholar
Rasiowa, H. 1974. An Algebraic Approach to Non-Classical Logics. North-Holland Publishing Company, Amster-dam. Google Scholar
Rasiowa, H., Sikorski, R. 1963. The Mathematics of Metamathematics. Polish Scientific Publishers (PWN), Warsaw. Google Scholar
Skowron, A., Jankowski, A., Dutta, S. 2025. Interactive Granular Computing: Toward Computing Model for Complex Intelligent Systems. In Proceedings of the 20th Conference on Computer Science and Intelligence Systems (FedCSIS), Bolanowski, M., Ganzha, M., Maciaszek, L., Paprzycki, M., Ślęzak, D. (Eds.).
Crossref
Google Scholar
Yao, Y.Y. 2004. Granular Computing: Basic Issues and Possible Solutions. In Proceedings of the 5th Joint Conference on Information Sciences. Google Scholar
Zadeh, L.A. 1965. Fuzzy Sets. Information and Control, 8(3), 338–353.
Crossref
Google Scholar
Zadeh, L.A. 1997. Toward a Theory of Fuzzy Information Granulation and Its Centrality in Human Reasoning and Fuzzy Logic. Fuzzy Sets and Systems, 90(2), 111–127.
Crossref
Google Scholar
a:1:{s:5:"en_US";s:3:"UWM";}
