On normal forms in Łukasiewicz logic (Q1882628): Difference between revisions

McNaughton functions of \(n\) variables are continuous, piecewise linear functions with integer coefficients from the hypercube \([0,1]^n\) to \([0,1]\). A McNaughton function is called simple, if there is a real polynomial \(g({\mathbf x}) = {\mathbf a \cdot \mathbf x } + b\) with integral coefficients such that \(f({\mathbf x})=g^\sharp ({\mathbf x})\) for every \({\mathbf x} \in [0,1]^n\) (where \(g^\sharp\) is defined on \([0,1]^n\) by \(g^\sharp({\mathbf x}) = 0 \vee (g({\mathbf x}) \wedge 1)\) as in \textit{R. L. O. Cignoli, I. L. M. D'Ottaviano} and \textit{D. Mundici}'s book [Algebraic foundations of many-valued reasoning. Dordrecht: Kluwer (2000; Zbl 0937.06009)]). In this paper an axiomatic description is given of some classes of formulas having the property to be canonically mapped one-to-one onto the class of simple McNaughton functions. Those classes are used to provide normal forms for Łukasiewicz sentential calculus and rational Łukasiewicz logic.