Singularity for graph-directed conjugate equations indexed by a two-vertex digraph
Kazuki Okamura · 2026
Read Singularity for graph-directed conjugate equations indexed by a two-vertex digraph by Kazuki Okamura (2026) 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-2603-21536.txt.
Introduction
Okamura studies graph-directed conjugate functional equations on the unit interval, indexed by a two-vertex digraph with self-loops. The focus is on when solutions are singular and when they are smooth. For compatible systems of weak contractions, Okamura first treats the affine case. Both solutions come out singular unless the two systems coincide. For a class of linear fractional systems, he proves a sharp dichotomy between singularity and smoothness. He also gives a sufficient condition for singularity in a nonlinear setting.
Catalog record
| CREATOR | Kazuki Okamura |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32644 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.21536
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Singularity for graph-directed conjugate equations indexed by a two-vertex digraph Kazuki Okamura Abstract . We study graph-directed conjugate functional equations on the unit interval indexed by the complete digraph with self-loops on two vertices. We focus on the singularity and regularity of the solutions for compatible systems of weak contractions. First, we show that both solutions are singular in the affine case unless the two systems coincide; second, we obtain a dichotomy between singularity and smoothness for a class of linear fractional systems; and finally, we give a sufficient condition for singularity in a non-linear setting. 2020 Mathematics Subject Classification 39B12, 39B72, 26A30 1. Introduction The de Rham functional equation belongs to an important class of iterative functional equations and has been studied by many authors ( [ 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 17 , 18 , 19 , 21 ] ). It can also be regarded as a special case of conjugate equations [ 3 , 14 ] . More recently, graph-directed versions of such equations have been considered in [ 11 ] , and [ 15 , Section 5] extended the framework on the unit interval to a graph-directed setting. In this paper, we deal with the simplest nontrivial graph-directed case, specifically, the complete digraph with self-loops on two vertices. Let \displaystyle\{f_{i,j}\}_{i,j\in\{0,1\}} and \displaystyle\{g_{i,j}\}_{i,j\in\{0,1\}} be transformations on \displaystyle[0,1] satisfying a certain compatibility condition. We consider the pair of solutions \displaystyle(\varphi_{0},\varphi_{1}) of g_{i,j}\circ\varphi_{j}=\varphi_{i}\circ f_{i,j},\ i,j\in\{0,1\}. (1.1) See Section 2 for the precise formulation. [ 15 , Section 5] establishes the existence and uniqueness of the pair of increasing and continuous solutions \displaystyle(\varphi_{0},\varphi_{1}) . In this paper, we consider the singularity and regularity of \displaystyle(\varphi_{0},\varphi_{1}) . It is a central theme in this area whether solutions are singular or smooth. Singularity for de Rham-type functional equations and, more generally, de Rham curves has been investigated. However, we are not aware of previous singularity results for graph-directed conjugate equations in the present framework. The main results of this paper are as follows. First, we prove that in the affine case both solutions \displaystyle\varphi_{0} and \displaystyle\varphi_{1} are singular unless the two affine systems \displaystyle\{f_{i,j}\}_{i,j} and \displaystyle\{g_{i,j}\}_{i,j} coincide, in which case both solutions are the identity. Second, we obtain a dichotomy between singularity and smoothness for a family with dyadic source maps \displaystyle\{f_{i,j}\}_{i,j} and linear fractional target maps \displaystyle\{g_{i,j}\}_{i,j} , and give explicit expressions for the solutions in the regular case. Third, we give a sufficient condition for singularity for a family with dyadic source maps \displaystyle\{f_{i,j}\}_{i,j} and non-linear target maps \displaystyle\{g_{i,j}\}_{i,j} in terms of the product of the Lipschitz constants. This paper is organized as follows. In Section 2 , we recall the framework from [ 15 , Section 5] . Section 3 considers the affine case. Section 4 studies the linear fractional case. Section 5 gives a singularity criterion in a non-linear setting. Finally, in Section 6 , we give some examples. 2. Framework We follow the setting of [ 15 , Section 5] . However, we use \displaystyle\{0,1\} instead of \displaystyle\{1,2\} as the index set for the maps. We adopt the Matkowski–Rus definition of weak contractions [ 9 , 16 ] . We say that a map \displaystyle\phi:[0,\infty)\to[0,\infty) is a comparison function if \displaystyle\phi is increasing and \displaystyle\displaystyle\lim_{n\to\infty}\phi^{n}(t)=0 for every \displaystyle t>0 . Let \displaystyle\phi be a comparison function. We say that a map \displaystyle h:[0,1]\to[0,1] is a \displaystyle\phi -contraction if \displaystyle|h(x)-h(y)|\leq\phi(|x-y|) for every \displaystyle x,y\in[0,1] . Since \displaystyle\phi(t)\leq t for every \displaystyle t\geq 0 , every \displaystyle\phi -contraction is continuous and its Lipschitz norm is less than or equal to \displaystyle 1 . We say that a doubly-indexed family of functions \displaystyle h_{i,j},i,j\in\{0,1\} , on \displaystyle[0,1] is a compatible system with a comparison function \displaystyle\phi if each \displaystyle h_{i,j} is strictly increasing, and a \displaystyle\phi -contraction, and furthermore a compatibility condition holds, specifically, 0=h_{i,0}(0)<h_{i,0}(1)=h_{i,1}(0)<h_{i,1}(1)=1,\ i\in\{0,1\}. (2.1) Let D^{h}_{i,n}\coloneqq\left\{h_{i,i_{1}}\circ h_{i_{1},i_{2}}\circ\cdots\circ h_{i_{n-1},i_{n}}(x_{n})\middle|i_{1},\dots,i_{n}\in\{0,1\},x_{n}\in\{0,1\}\right\} and \displaystyle\displaystyle D^{h}_{i}\coloneqq\bigcup_{n\geq 1}D^{h}_{i,n} . Then \displaystyle(D^{h}_{i,n})_{n} is increasing with respect to \displaystyle n . Let I^{h}_{i}(i_{1},\dots,i_{n})\coloneqq\left[h_{i,i_{1}}\circ h_{i_{1},i_{2}}\circ\cdots\circ h_{i_{n-1},i_{n}}(0),h_{i,i_{1}}\circ h_{i_{1},i_{2}}\circ\cdots\circ h_{i_{n-1},i_{n}}(1)\right]. By the compatibility condition ( 2.1 ), I^{h}_{i}(i_{1},\dots,i_{n})=I^{h}_{i}(i_{1},\dots,i_{n},0)\cup I^{h}_{i}(i_{1},\dots,i_{n},1). (2.2) For every \displaystyle x\in[0,1] , there exists a sequence \displaystyle(i_{n})_{n} in \displaystyle\{0,1\} such that \displaystyle x\in I^{h}_{i}(i_{1},\dots,i_{n}) for each \displaystyle n , and, if \displaystyle x\in[0,1]\setminus D^{h}_{i} , then such a sequence is unique. We can choose \displaystyle i_{n} inductively. Indeed, by ( 2.2 ), if \displaystyle x\in I^{h}_{i}(i_{1},\dots,i_{n}) , then \displaystyle x\in I^{h}_{i}(i_{1},\dots,i_{n},0) or \displaystyle x\in I^{h}_{i}(i_{1},\dots,i_{n},1) . If additionally \displaystyle x\in[0,1]\setminus D^{h}_{i} , then exactly one of \displaystyle x\in I^{h}_{i}(i_{1},\dots,i_{n},0) and \displaystyle x\in I^{h}_{i}(i_{1},\dots,i_{n},1) holds. We denote the length of an interval \displaystyle I by \displaystyle|I| . Let \Delta^{h}_{i,n}\coloneqq\max_{i_{1},\dots,i_{n}\in\{0,1\}}\left|I^{h}_{i}(i_{1},\dots,i_{n})\right|,\ n\geq 1,\ i\in\{0,1\}. Since \displaystyle\left|I^{h}_{i}(i_{1},\dots,i_{n})\right|\leq\phi^{n}(1) and \displaystyle\phi is a comparison function, \lim_{n\to\infty}\Delta^{h}_{i,n}=0. (2.3) Hence, \displaystyle D^{h}_{i} is dense in \displaystyle[0,1] . Let \displaystyle\{f_{i,j}\}_{i,j} and \displaystyle\{g_{i,j}\}_{i,j} be two compatible systems on \displaystyle[0,1] with a common comparison function \displaystyle\phi under the Euclidean distance. For each \displaystyle i\in\{0,1\} , we define functions \displaystyle\varphi_{i} , on \displaystyle D^{f}_{i} such that for \displaystyle i_{1},\dots,i_{n}\in\{0,1\},x_{n}\in\{0,1\} , \varphi_{i}\left(f_{i,i_{1}}\circ f_{i_{1},i_{2}}\circ\cdots\circ f_{i_{n-1},i_{n}}(x_{n})\right)=g_{i,i_{1}}\circ g_{i_{1},i_{2}}\circ\cdots\circ g_{i_{n-1},i_{n}}(x_{n}). This is well-defined by arguments in [ 15 , Section 5] . Now we define \displaystyle\varphi_{i}(x) for \displaystyle x\in[0,1]\setminus D^{f}_{i} . Take a unique sequence \displaystyle(i_{n})_{n} in \displaystyle\{0,1\} such that \displaystyle x\in I^{f}_{i}(i_{1},\dots,i_{n}) for every \displaystyle n\geq 1 . Then, \displaystyle\displaystyle\bigcap_{n\geq 1}I^{g}_{i}(i_{1},\dots,i_{n}) is a compact subset of \displaystyle[0,1] and by ( 2.3 ), it is a singleton. Let \displaystyle y\in[0,1] such that \displaystyle\displaystyle\{y\}=\bigcap_{n\geq 1}I^{g}_{i}(i_{1},\dots,i_{n}) . Then we define \displaystyle\varphi_{i}(x)\coloneqq y . Thus a map \displaystyle\varphi_{i}:[0,1]\to[0,1] is defined. By [ 15 , Proposition 5.1] , each \displaystyle\varphi_{i} is increasing and continuous on \displaystyle[0,1] . The pair of maps \displaystyle(\varphi_{0},\varphi_{1}) satisfies the following system of functional equations: \begin{cases}g_{0,0}(\varphi_{0}(x))=\varphi_{0}(f_{0,0}(x))\\ g_{0,1}(\varphi_{1}(x))=\varphi_{0}(f_{0,1}(x))\\ g_{1,0}(\varphi_{0}(x))=\varphi_{1}(f_{1,0}(x))\\ g_{1,1}(\varphi_{1}(x))=\varphi_{1}(f_{1,1}(x))\end{cases}x\in[0,1]. (2.4) The main purpose of this paper is to consider the regularity and singularity of the solutions \displaystyle\varphi_{0} and \displaystyle\varphi_{1} in ( 2.4 ). We can also define a compatible system for a single-indexed family in the same manner; see [ 12 , Subsection 2.1] . Assume that \displaystyle\{f_{0},f_{1}\} and \displaystyle\{g_{0},g_{1}\} are both compatible systems and \displaystyle f_{i,j}=f_{j} and \displaystyle g_{i,j}=g_{j} for every \displaystyle i,j\in\{0,1\} . Then, \displaystyle\varphi_{0}=\varphi_{1} on \displaystyle[0,1] and ( 2.4 ) is equivalent to \begin{cases}g_{0}(\varphi(x))=\varphi(f_{0}(x))\\ g_{1}(\varphi(x))=\varphi(f_{1}(x))\end{cases}x\in[0,1]. (2.5) This is the conjugate equation studied in [ 14 ] , which contains the framework of de Rham functional equations on \displaystyle[0,1] . We denote the Lebesgue measure by \displaystyle\ell . We say that a function \displaystyle\varphi on \displaystyle[0,1] is singular if \displaystyle\varphi^{\prime}(x)=0 , \displaystyle\ell -a.e. \displaystyle x\in(0,1) . We recall the monotone differentiation theorem, which states that every increasing function on \displaystyle[0,1] is Lebesgue measurable and differentiable almost everywhere with respect to \displaystyle\ell . See [ 20 , Theorem 1.6.25] for more details. Finally, we state a connection with graph-directed invariant sets. Graph-directed invariant sets were first introduced by Mauldin and Williams [ 10 ] as a generalization of Hutchinson’s self-similar sets. Let \displaystyle V\coloneqq\{0,1\} . For \displaystyle i,j\in V , let \displaystyle e_{i,j} be a directed edge from \displaystyle i to \displaystyle j . We remark that \displaystyle i=j is allowed and there are exactly four directed edges. See Figure 1 below. \displaystyle 0 \displaystyle 1 \displaystyle e_{0,1} \displaystyle e_{1,0} \displaystyle e_{0,0} \displaystyle e_{1,1} Figure 1. two-vertex complete digraph with