On the Non-Uniqueness of Representation of ( U , N ) (U,N) -Implications Journal: Fuzzy Sets and Systems
Raquel Fernandez-Peralta, Andrea Mesiarová-Zemánková · 2025
Read On the Non-Uniqueness of Representation of ( U , N ) (U,N) -Implications Journal: Fuzzy Sets and Systems by Raquel Fernandez-Peralta, Andrea Mesiarová-Zemánková (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-16299.txt.
Introduction
In fuzzy logic, one family of implication rules is built from a disjunctive uninorm plus a fuzzy negation. Raquel Fernandez-Peralta and Andrea Mesiarová-Zemánková ask whether that building recipe is unique for each rule. Prior work on a close family suggested a unique recipe when the negation is smooth, and that belief carried over.
The pair show the belief is false: such rules can have more than one valid recipe, even with a smooth negation. They also study when uniqueness does hold, for both smooth and rough building blocks. The result sharpens the theory of how these operators are put together.
Catalog record
| CREATOR | Raquel Fernandez-Peralta, Andrea Mesiarová-Zemánková |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 67 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-16299
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On the Non-Uniqueness of Representation of ( U , N ) (U,N) -Implications Journal: Fuzzy Sets and Systems Raquel Fernandez-Peralta, Andrea Mesiarová-Zemánková Abstract Fuzzy implication functions constitute fundamental operators in fuzzy logic systems, extending classical conditionals to manage uncertainty in logical inference. Among the extensive families of these operators, generalizations of the classical material implication have received considerable theoretical attention, particularly ( S , N ) (S,N) -implications constructed from t-conorms and fuzzy negations, and their further generalizations to ( U , N ) (U,N) -implications using disjunctive uninorms. Prior work has established characterization theorems for these families under the assumption that the fuzzy negation N N is continuous, ensuring uniqueness of representation. In this paper, we disprove this last fact for ( U , N ) (U,N) -implications and we show that they do not necessarily possess a unique representation, even if the fuzzy negation is continuous. Further, we provide a comprehensive study of uniqueness conditions for both uninorms with continuous and non-continuous underlying functions. Our results offer important theoretical insights into the structural properties of these operators. 1 Introduction Fuzzy implication functions serve as fundamental operators in fuzzy logic, extending classical logical conditionals to accommodate the continuous nature of approximate reasoning. These functions have become pivotal in computational intelligence due to their ability to model uncertainty-driven inference [ 2 ] , finding applications across diverse domains including image processing [ 8 ] , knowledge discovery [ 14 , 15 ] , and intelligent control systems [ 25 ] . Beyond their practical utility, the theoretical study of these operators is an important research area encompassing the development of new families of fuzzy implication functions, the analysis of their additional properties, and the pursuit of structural representations and characterizations. Indeed, numerous families of these operators have been introduced in the literature under different motivations, along with various additional properties [ 13 ] . This proliferation of operators has underscored the critical importance of studying their characterization, examining their intersections, and revisiting long-standing open problems in the field [ 24 ] . Among the vast array of fuzzy implications functions, generalizations of the classical material implication p\to q\equiv\neg q\vee p represent one of the most fundamental and thoroughly investigated classes. Historically, these generalizations first emerged in the form of S -implications, constructed using t-conorms and strong negations [ 33 ] . Subsequent developments expanded this framework to (S,N) -implications by incorporating arbitrary fuzzy negations [ 20 ] , and further generalized to (A,N) -implications where aggregation functions replace t-conorms [ 29 ] . A particularly notable subclass arises when disjunctive uninorms are used, leading to (U,N) -implications [ 7 ] , which is a family that is suitable to define a fuzzy morphology based on uninorms [ 17 ] . Throughout the years, there have been several advances in the characterization of these families of fuzzy implication functions: In [ 1 ] , the authors characterized (S,N) -implications with continuous negations, while [ 4 , 22 , 35 ] extended these results to uninorms, 2-uninorms and grouping functions under the same continuity condition. Later, [ 12 ] provided a representation theorem of (S,N) -implications for non-continuous negations, linking the problem to the completion of t-conorms in undefined regions [ 10 , 11 ] . A key step in these results is the reconstruction of the aggregation function from the given negation and implication. Depending on the conditions, this reconstruction may or may not be unique—a direct consequence of whether the fuzzy implication’s representation in terms of the aggregation function is unique. For example, in the case of (S,N) -implications, uniqueness holds when the negation N is continuous but fails otherwise. Similarly, [ 4 ] claimed that (U,N) -implications have a unique representation when the negation N is continuous. However, in this work we prove that (U,N) -implications do not generally possess a unique representation, even when N is continuous. This finding raises new questions regarding the conditions that guarantee uniqueness of representation, as well as the characterization of all uninorms generating the same fuzzy implication function. In this paper, we provide a comprehensive investigation of both problems, considering uninorms with both continuous and non-continuous underlying functions [ 27 , 28 ] . The structure of the paper is as follows. First, in Section 2 we include the necessary background. In Section 3 we prove that the representation of (U,N) -implications is not unique. In Section 4 we present the results regarding the uniqueness of representation. The paper ends in Section 5 with some conclusions and future work. 2 Preliminaries In this section, we provide the results and definitions necessary for making this paper self-contained. For more information regarding fuzzy implications and aggregation functions we refer the reader to the books [ 2 , 5 , 6 , 18 , 19 ] . A fuzzy negation N is a decreasing unary function which is the generalization of the classical negation. Definition 2.1 ( [ 16 ] ). A unary function N:[0,1]\to[0,1] is said to be a fuzzy negation if it satisfies: (N1) N(0)=1 and N(1)=0 . (Boundary Condition) (N2) N is decreasing. (Monotonicity) A fuzzy negation N is called (i) strict , if it is strictly decreasing and continuous. (ii) strong , if it is an involution, i.e., N(N(x))=x for all x\in[0,1] . It is straightforward to prove that any strong fuzzy negation is strict. For some results, it is useful to consider the pseudo-inverse of fuzzy negations which is defined as follows. Definition 2.2 ( [ 1 ] ). Let N be a fuzzy negation, the modified pseudo-inverse of N is denoted by \mathfrak{R}_{N} and it is defined as follows \mathfrak{R}_{N}(x)=\left\{\begin{array}[]{ll}1&\text{if }x=0,\\ N^{(-1)}(x)&\text{if }x\in]0,1],\\ \end{array}\right. (1) where N^{(-1)} is the pseudo-inverse of N given by N^{(-1)}(x)=\sup\{y\in[0,1]\,|\,N(y)>x\},\quad\text{for all }x\in[0,1]. The pseudo-inverse of a continuous negation is a strictly decreasing fuzzy negation that satisfies the inverse function equalities, but one of them only when restricted to its range. Proposition 2.1 ( [ 1 , Proposition 3.13] ). If N is a continuous fuzzy negation, then \mathfrak{R}_{N} is a strictly decreasing fuzzy negation. Moreover, the following statements hold: (i) \mathfrak{R}_{N}^{(-1)}=N . (ii) N\circ\mathfrak{R}_{N}=\text{id}_{[0,1]} . (iii) \mathfrak{R}_{N}\circ N\mid_{\text{Ran }\mathfrak{R}_{N}}=\text{id}_{\text{Ran }\mathfrak{R}_{N}} . Bivariate aggregation functions were originally developed as generalizations of classical conjunctions and disjunctions. Definition 2.3 ( [ 30 , Definition 3] ). A bivariate aggregation function is an increasing function A:[0,1]^{2}\to[0,1] verifying the boundary conditions A(0,0)=0 and A(1,1)=1 . Since in this work we only deal with two variable aggregation functions, we omit the term bivariate for simplification. An aggregation function is called conjuctive if A\leq\min and disjunctive if A\geq\max . Among the most important aggregation functions are uninorms, characterized by associativity and the existence of a neutral element e\in[0,1] . Definition 2.4 ( [ 19 , 34 ] ). A binary operator U:[0,1]^{2}\to[0,1] is said to be a uninorm if there exists e\in[0,1] , called neutral element , such that U satisfies: (U1) U(x,y)=U(y,x) for all x,y\in[0,1] . (Commutativity) (U2) U(x,U(y,z))=U(U(x,y),z) for all x,y,z\in[0,1] . (Associativity) (U3) U(x,y)\leq U(x,z) when y\leq z , for all x,y,z\in[0,1] . (Monotonicity) (U4) U(x,e)=x for all x\in[0,1] . (Neutral element) For all uninorms, it holds that U(1,0)\in\{0,1\} and we say that U is conjunctive if U(1,0)=0