On the Algebraic Independence of E E - and G G -Functions, II: An Effective Version
\firstname Daniel \middlename \lastname vargas-Montoya · 2025
Read On the Algebraic Independence of E E - and G G -Functions, II: An Effective Version by \firstname Daniel \middlename \lastname vargas-Montoya (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-2507-20429.txt.
Introduction
Daniel Vargas-Montoya proves an effective test for when special power series stand algebraically apart in a 2025 paper. The series live over a p-adic field and solve differential equations with strong Frobenius structure. A maximal order condition at zero trims the class further.
That class turns out to hold a rich family of E-functions and G-functions, two storied kinds of series from number theory. Vargas-Montoya gives a criterion that decides algebraic independence over the field of analytic elements for members of the class. Worked examples then show concrete E- and G-functions passing the test. Results of this kind matter because independence proofs show that no hidden polynomial relation ties the values together, a basic fact about how new these numbers are.
Catalog record
| CREATOR | \firstname Daniel \middlename \lastname vargas-Montoya |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 107 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-20429
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On the Algebraic Independence of E E - and G G -Functions, II: An Effective Version \firstname Daniel \middlename \lastname vargas-Montoya Abstract . Let K K be a finite extension of ℚ p \mathbb{Q}_{p} that is totally ramified over ℚ p \mathbb{Q}_{p} . The set ℳ ℱ ( K ) \mathcal{M}\mathcal{F}(K) consists of power series in 1 + z K [ [ z ] ] 1+zK[[z]] that are solutions of differential operators in K ( z ) [ d / d z ] K(z)[d/dz] equipped with strong Frobenius structure and satisfying maximal order multiplicty (MOM) condition at zero. It turns out that this set contains an interesting class of E E - and G G -functions. In this work, we provide a criterion for determining the algebraic independence, over the field of analytic elements, of elements belonging to ℳ ℱ ( K ) \mathcal{M}\mathcal{F}(K) . As an illustration of this criterion, we show the algebraic independence of some E E - and G G -functions over the field of analytic elements. \alttitle \altkeywords 1. Introduction In [ 12 ] , Siegel introduced the class of E - and G -functions. His main purpose in introducing these classes was to generalise the classical theorems of Hermite and Lindemann-Weierstrass . Recall that a power series f(z)=\sum_{n\geq 0}a_{n}z^{n} is a G - function if the coefficients a_{n} are algebraic numbers and there is a real number C>0 such that: 1. the power series f(z) is solution of a nonzero differential operator with coefficients in \overline{\mathbb{Q}}(z) ; 2. the absolute values of all Galois conjugates of a_{n} are at most C^{n+1} for all n\geq 0 ; 3. there is a sequence of positive integers D_{m} such that, for all integers m\geq 0 , |D_{m}|<C^{m+1} and, for all n\leq m , D_{m}a_{n} is an algebraic integer. Among the G -functions, we have the hypergeometric series {}_{n}F_{n-1} with rational parameters, and diagonal of rational functions . Furthermore, a power series F(z)=\sum_{n\geq 0}\frac{a_{n}}{n!}z^{n} is an E - function if and only if f(z)=\sum_{n\geq 0}a_{n}z^{n} is a G -function. In particular, this implies that if F(z) is an E -function, then it is a solution of a nonzero differential operator with coefficients in \overline{\mathbb{Q}}(z) . Classical examples of E -functions include the exponential function, sine, cosine, hypergeometric series {}_{p}F_{p} with rational parameters, and the Bessel function. Siegel developed a method for proving the algebraic independence of values of E -functions at algebraic points. He also suggested that his method could be applied to G -functions, and this was later accomplished by Chudnovsky [ 7 ] in 1984. These results have motivated the study of algebraic independence of E - and G -functions, and since then, numerous results have been established. For instance, in [ 4 ] and [ 13 ] , the authors established criteria for the algebraic independence of E -functions. In the case of G -functions, the works [ 2 ] and [ 17 ] proved the transcendence of certain G -functions that are p -Lucas for infinitely many primes p , and recently, in [ 1 ] , a criterion for the algebraic independence of such G -functions was given. The present work is motivated by a recent result of [ 16 ] , where we established a criterion for the algebraic independence of power series in \mathcal{M}\mathcal{F}(K) with K a Frobenius field. Recall that K is a Frobenius field if K is a finite extension of \mathbb{Q}_{p} and, for all x\in K such that |x|\leq 1 , we have |x-x^{p}|<1 . Examples of Frobenius fields are given by finite extensions of \mathbb{Q}_{p} that are totally ramified over \mathbb{Q}_{p} . In order to recall the definition of the set \mathcal{M}\mathcal{F}(K) , let us first review the definition of the field of analytic elements . Let \mathbb{C}_{p} be the completion of \overline{\mathbb{Q}_{p}} with respect to the p -adic norm and let L be a complete extension of \mathbb{Q}_{p} such that L\subset\mathbb{C}_{p} . The field L(z) is equipped with the Gauss norm \left|\frac{P(z)}{Q(z)}\right|_{\mathcal{G}}=\frac{\max\{|a_{i}|\}_{1\leq i\leq n}}{\max\{|b_{j}|\}_{1\leq j\leq n}}, where P(z)=\sum a_{i}z^{i} and Q(z)=\sum b_{j}z^{j} belong to L(z) . Further, for a power series f(z)=\sum_{n\geq 0}a_{n}z^{n}\in L[[z]] with |a_{n}|\leq 1 for all n\geq 0 , we set |f(z)|_{\mathcal{G}}=\sup_{n\geq 0}|a_{n}| . The completion of L(z) with respect to the Gauss norm is called the field of analytic elements and is denoted by E_{L} . When L=\mathbb{C}_{p} , the field E_{\mathbb{C}_{p}} is denoted E_{p} . The set \mathcal{M}\mathcal{F}(K) consists of power series f(z) in 1+zK[[z]] such that f(z) is solution of a differential operator \mathcal{L}\in E_{p}[d/dz] which is MOM (maximal order multiplicity) at zero and has strong Frobenius structure. For the precise definition of MOM and strong Frobenius structure, we refer the reader to [ 16 , Section 2.2] . Nevertheless, let us point out that if \mathcal{L}\in\mathbb{Q}(z)[d/dz] , being MOM at zero is equivalent to having maximal unipotent monodromy (MUM) at zero. Moreover, the strong Frobenius structure can be viewed as a Frobenius action on the solution space of \mathcal{L} . As highlighted in [ 16 , Section 2.4] , \mathcal{M}\mathcal{F}(K) contains an interesting class of E - and G -functions. For example, for any prime number p>2 , the classical G -functions \mathfrak{f}_{r}(z)=\sum_{n\geq 0}\binom{2n}{n}^{r}z^{n},\text{ }\quad\mathfrak{A}(z)=\sum_{n\geq 0}\left(\sum_{k=0}^{n}\binom{n}{k}^{2}\binom{n+k}{k}^{2}\right)z^{n} belong to \mathcal{M}\mathcal{F}(\mathbb{Q}_{p}) . Note that \mathfrak{f}_{r}(z) is the hypergeometric series {}_{r}F_{r-1}((1/2,\ldots,1/2),(1,\ldots,1),4^{r}z) and \mathfrak{A}(z) is the Apéry power series. Also, classical examples of E -functions, such as the exponential and Bessel functions, given respectively by \exp(\pi_{p}z)=\sum_{n\geq 0}\frac{\pi_{p}^{n}}{n!}z^{n},\text{}\quad J_{0}(\pi_{p}z)=\sum_{n\geq 0}\frac{(-1)^{n}\pi_{p}^{2n}}{4^{n}(n!)^{2}}z^{2n}, belong to \mathcal{M}\mathcal{F}(K_{p}) , where K_{p}=\mathbb{Q}_{p}(\pi_{p}) and \pi_{p}^{p-1}=-p . Further, inside E_{L} , we have the ring E_{0,L} , which is the completion of L_{0}(z) , the ring of rational functions A(z)/B(z)\in L(z) such that B(x)\neq 0 for all x\in D_{0} , where D_{0}=\{x\in\mathbb{C}_{p}:|x|<1\} . The ring E_{0,\mathbb{C}_{p}} will be denoted by E_{0,p} . Note that E_{0,L}\subset L[[z]] . By combining differential Galois theory and p -adic differential equations, we established in [ 16 ] the following result. {theo} [Theorems 2.3 and 2.4 of [ 16 ] ] Let K be a Frobenius field and let f_{1}(z),\ldots,f_{m}(z)\in\mathcal{M}\mathcal{F}(K) . Then: (1) f_{1}(z),\ldots,f_{m}(z) are algebraically dependent over E_{K} if and only if there exist integers a_{1},\ldots,a_{m}\in\mathbb{Z} , not all zero, such that f(z)^{a_{1}}_{1}\cdots f(z)^{a_{m}}_{m}\in E_{0,K}; (2) f_{1}(z),\ldots,f_{m}(z) are algebraically dependent over E_{K} if and only if, for any (r_{1},\ldots r_{m})\in\mathbb{N}^{m} , f_{1}^{(r_{1})},\ldots,f_{m}^{(r_{m})} are algebraically dependent over E_{K} . Inspired by this result, the goal of the present work is to study the transcendence and algebraic independence over E_{K} of E - and G -functions that lie in \mathcal{M}\mathcal{F}(K) . A first illustration of Theorem 1 is given in Section 2 , where we prove a conjecture due to Christol [ 6 , p.30] , asserting that J_{0}(\pi_{p}z) is transcendental over E_{K_{p}} for p>2 . It is important to note that \exp(\pi_{p}z) is algebraic over E_{K_{p}} because \exp(\pi_{p}z)^{p}\in E_{K_{p}} . However, in order to address the question of algebraic independence, it becomes necessary to establish a general criterion for determining when, for power series f_{1}(z),\ldots,f_{m}(z)\in\mathcal{M}\mathcal{F}(K) , the expression f_{1}(z)^{a_{1}}\cdots f_{m}(z)^{a_{m}}\notin E_{K} for any nonzero tuple (a_{1},\ldots,a_{m})\in\mathbb{Z}^{m} . The formulation of this criterion requires recalling the notions of poles and residues for elements in E_{0,K} , as well as