Infinite convex geometries with lower semi-modularity and join semi-distributivity
Adam Mata · 2025
Read Infinite convex geometries with lower semi-modularity and join semi-distributivity by Adam Mata (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-2508-15511.txt.
Introduction
Adam Mata studies infinite convex geometries in this 2025 math paper. A convex geometry is a lattice that mimics how convex sets close up in space; think of adding points until a shape is filled. Two order traits matter here: lower semi-modularity and join semi-distributivity. In infinite settings these traits can fail even when every finite piece has them.
Mata defines a class of infinite geometries built by stacking finite ones along a chain and taking the union. He proves that members of this class keep both traits, which do not come for free at infinity. He also shows the chain method cannot build every infinite convex geometry, so the class is a proper part of the whole. The paper serves lattice theorists who need to know which finite-world rules survive the move to infinity.
Catalog record
| CREATOR | Adam Mata |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 22 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-15511
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Infinite convex geometries with lower semi-modularity and join semi-distributivity Adam Mata Abstract . The following article treats about convex geometries which are lower semi-modular and join semi-distributive lattices. Firstly, it is shown that there is a class K K of infinite convex geometries which can be build out of finite ones by using the construction of a union of a chain. Then it is shown that elements of K K preserve lower semi-modularity and join semi-distributivity which are not default properties in the infinite setting. It is also discussed that not all of the infinite convex geometries may be obtained by the means of the union of a chain. 2020 Mathematics Subject Classification 03Cxx, 06Bxx 1. Introduction Convex geometries first occurred in [ 9 ] , discovered as finite lower semimodular lattices with a unique irredundant decomposition. Their study as closure operators with the anti-exchange property is carried out in [ 10 ] . Since then they have also been studied as lattices in [ 1 ] and [ 2 ] . In [ 6 ] the authors carried out the research concerning SPS lattices which turn out to be dual to convex geometries of convex dimension 2. As long as they characterized the congruences of SPS lattices hence we also obrained this classification for cdim=2 . The general survey on the topic of convex geometries may be found in [ 11 , Chapter 5] . Convex geometries are also duals to antimatroids (see [ 8 ] ) and a special type of greedoids (see [ 13 ] ). The infinite case was presented and studied in both [ 1 ] and [ 3 ] . It turns out that convex geometries (also infinite) also found their application in conditional logic as presented in [ 15 ] and [ 14 ] . In this paper, we show how to build an infinite convex geometry out of the finite ones and reveal them as an example of class of structures which has the properties of lower semi-modularity and join semi-distributivity . This setting of properties is not common in the case of infinite convex geometries as it was shown in both, accordingly, [ 1 , Section 1] and [ 3 , Theorem 10] . We start the paper with some preliminaries concerning convex geometries and its representations as well as some model-theoretic tools. Then we show how to define convex geometries by \forall\exists formulas. Later we present the construction of infinite convex out of finite ones as well as the fact that not all finite convex geometries are elementary equivalent. We finish with the construction of infinite convex geometries which are (JSD) and (LSM). 2. Preliminaries Definition 2.1 . Let X be a nonempty set. We say that mapping \alpha:2^{X}\rightarrow 2^{X} is an algebraic closure operator if, for all A,B\subseteq X : • A\subseteq\alpha(A) , • A\subseteq B\implies\alpha(A)\subseteq\alpha(B) , • \alpha(\alpha(A))=\alpha(A) , • \alpha(\varnothing)=\varnothing . Definition 2.2 . ( [ 10 , Section 2] ) A convex geometry is a pair \mathcal{G}=(X,\alpha) where: • X is a nonempty set • \alpha is an algebraic closure operator with the anti-exchange property , i. e.: (AE) q\in\alpha(K\cup\{p\})\implies p\not\in\alpha(K\cup\{q\}) where K\subseteq X is \alpha-closed , p,q\in X but p,q\not\in K . In general, we associate \alpha with the set of all closed sets obtained by \alpha , which is \{\alpha(Y):\ Y\subseteq X\} . The above definition is equivalent to the following. Definition 2.3 . ( [ 14 , Section 2.1] ) A convex geometry is a pair \mathcal{G}=(X,\alpha) where X is a nonempty set and \alpha is a family of subsets of X such that: • \alpha is closed under intersection , i. e. for any \beta\subseteq\alpha it holds: \bigcap\beta\in\alpha • \alpha has an anti-exchange property , i. e. for every A\in\alpha and every x,y\in X with x,y\in X and x\not=y there is B\in\alpha with A\subseteq B such that x\in B and y\not\in B , or x\not\in B and y\in B . In case of finite convex geometries, a set of closed sets under intersection may be also introduced by an idea of downsets of finite generating chains. This notion is introduced and developed in [ 10 ] . It was proven in this paper, that any finite convex geometry may be obtained by the following construction: Let’s take a finite set X=\{x_{1},x_{2},\dots x_{n}\} . We take K different chains composed out of elements from X : • C_{1}=x_{i^{1}_{1}}<x_{i^{1}_{2}}<\dots<x_{i^{1}_{n}} , • C_{2}=x_{i^{2}_{1}}<x_{i^{2}_{2}}<\dots<x_{i^{2}_{n}} , • \dots • C_{K}=x_{i^{K}_{1}}<x_{i^{K}_{2}}<\dots<x_{i^{K}_{n}} . Let \mathfrak{D}(C_{i}) denote a family of downsets of the chain C_{i} . For example: (1) \mathfrak{D}(C_{1})=\{\varnothing,\{x_{i^{1}_{1}}\},\{x_{i^{1}_{1}},x_{i^{1}_{2}}\},\dots,X\} Set \mathfrak{D}=\bigcup_{i=1}^{K}\mathfrak{D}(C_{i}) . Let \mathfrak{D}^{\cap} be a set \mathfrak{D} closed under intersection. Then \mathcal{G}=(X,\mathfrak{D}^{\cap}) is a convex geometry. A minimal number of chains needed to generate a particular convex geometry is called a convex dimension and is denoted as cdim(\mathcal{G}) . We exploit this schema of construction further in the paper. 3. Universal and \forall\exists Formulas - Preservation by a Union of a Chain Definition 3.1 . A universal formula is a formula in a form: \forall x_{1}\forall x_{2}\dots\forall x_{n}:A(x_{1},x_{2},\dots x_{n}), where A(x_{1},x_{2},\dots x_{n}) is a formula which does not contain any quantifier and n may be equal to 0 . Further, a \forall\exists formula is formula in the form: \forall x_{1}\forall x_{2}\dots\forall x_{n}\exists y_{1}\exists y_{2}\dots\exists y_{m}:A(x_{1},x_{2},\dots x_{n},y_{1},y_{2},\dots y_{m}), where A(x_{1},x_{2},\dots x_{n},y_{1},y_{2},\dots y_{m}) is a formula which does not contain any quantifier and n or m (or both) may be equal to 0 . Let us notice that every universal formula is an instance of a \forall\exists formula. Further, we introduce an idea of a union of a chain of models. The following facts may be found in [ 7 , Chapter 8] . 3.1. A union of a chain of models Definition 3.2 . [ 7 , Definition 8.1] Let L be a first order language, let \mathcal{M} be an L -structure and let \mathcal{N} be a substructure of \mathcal{M} . We say that \mathcal{N} is an elementary substructure of \mathcal{M} if, for every formula F(x_{1},x_{2},\dots x_{k}) and elements n_{1},n_{2},\dots n_{k}\in N we have: \mathcal{M}\models F(n_{1},n_{2},\dots n_{k})\iff\mathcal{N}\models F(n_{1},n_{2},\dots n_{k}). Definition 3.3 . [ 7 , Definition 8.10] A function h:M\rightarrow N is an elementary mapping if, for every formula F(x_{1},x_{2},\dots x_{k}) and elements m_{1},m_{2},\dots m_{k}\in M we have: \mathcal{M}\models F(m_{1},m_{2},\dots m_{k})\iff\mathcal{N}\models F(h(m_{1}),h(m_{2}),\dots h(m_{k})). In such a case we say that \mathcal{M} can be elementary embedded into \mathcal{N} . Remarkable observation is that any elementary mapping is in fact an embedding. It is sufficient to consider a formula A(x_{1},x_{2})=x_{1}\simeq x_{2} to notice so. Let h be a monomorphism from \mathcal{M} into \mathcal{N} . It is an elementary mapping if and only if the image of h is an elementary substructure of \mathcal{N} . Let \mathcal{M} be an L -structure. The complete theory of \mathcal{M} , denoted by Th(\mathcal{M}) , is a set of formulas: Th(\mathcal{M})=\{\phi:\phi\text{ is a closed formula of $L$ and }\mathcal{M}\models\phi\}. If \mathcal{M} and \mathcal{N} are both L -structures, we say that \mathcal{M} and \mathcal{N} are elementary equivalent if Th(\mathcal{M})=Th(\mathcal{N}) , what is denoted by \mathcal{M}\equiv\mathcal{N} . The following theorem gives a tool to discover whether two structures are elementary equivalent. Theorem 3.4 . [ 7 , Theorem 8.22] Let \mathcal{M}_{1} and \mathcal{M}_{2} be L -structures. \mathcal{M}_{1}\equiv\mathcal{M}_{2} if