Algebraic independence of Mahler functions via radial asymptotics

  title={Algebraic independence of Mahler functions via radial asymptotics},
  author={Richard P. Brent and Michael Coons and Wadim Zudilin},
  journal={arXiv: Number Theory},
We present a new method for algebraic independence results in the context of Mahler's method. In particular, our method uses the asymptotic behaviour of a Mahler function $f(z)$ as $z$ goes radially to a root of unity to deduce algebraic independence results about the values of $f(z)$ at algebraic numbers. We apply our method to the canonical example of a degree two Mahler function; that is, we apply it to $F(z)$, the power series solution to the functional equation $F(z)-(1+z+z^2)F(z^4)+z^4F(z… 

Figures and Tables from this paper

Becker's conjecture on Mahler functions

Becker's conjecture is proved in the best-possible form; it is shown that the rational function R(z) can be taken to be a polynomial for some explicit non-negative integer $\gamma$ and such that $1/Q (z)$ is $k$-regular.

An asymptotic approach in Mahler's method

We provide a general result for the algebraic independence of Mahler functions by a new method based on asymptotic analysis. As a consequence of our method, these results hold not only over

Transcendence tests for Mahler functions

We give two tests for transcendence of Mahler functions. For our first, we introduce the notion of the eigenvalue $\lambda_F$ of a Mahler function $F(z)$, and develop a quick test for the

Méthode de Mahler : relations linéaires, transcendance et applications aux nombres automatiques

This paper is devoted to the so‐called Mahler method. We precisely describe the structure of linear relations between values at algebraic points of Mahler functions. Given a number field k , a Mahler

Hypertranscendence and algebraic independence of certain infinite products

a(z)b(z) = p(z)b(z)a(z). Since a(z) and b(z) are coprime, it follows that a(z) | a(z) giving a(z) = c ∈ C×. Thus p(z) = b(z)/b(z). On the other hand, if p(z) is of this form, then F (z) = 1/b(z) is a

A sequential view of self--similar measures, or, What the ghosts of Mahler and Cantor can teach us about dimension

We show that missing $q$-ary digit sets $F\subseteq[0,1]$ have corresponding naturally associated countable binary $q$-automatic sequence $f$. Using this correspondence, we show that the Hausdorff

Mahler Takes a Regular View of Zaremba

This paper associates a regular sequence---in the sense of Allouche and Shallit---and establishes various properties and results concerning the generating function of the regular sequence.

On linear independence measures of the values of Mahler functions

  • K. VäänänenWen Wu
  • Mathematics
    Proceedings of the Royal Society of Edinburgh: Section A Mathematics
  • 2018
We estimate the linear independence measures for the values of a class of Mahler functions of degrees 1 and 2. For this purpose, we study the determinants of suitable Hermite–Padé approximation

Mahler ’ s Method

α−4 + 1 α−8 + · · · and with the algebraic independence of the numbers f(α), f′(α), f”(α), . . .. Here, α denotes again an algebraic number with 0 < |α| < 1. Moreover, examples of this kind can be

A Density Theorem for the Difference Galois Groups of Regular Singular Mahler Equations

  • Marina Poulet
  • Mathematics
    International Mathematics Research Notices
  • 2021
The difference Galois theory of Mahler equations is an active research area. The present paper aims at developing the analytic aspects of this theory. We first attach a pair of connection matrices



Algebraic Independence of Infinite Products and Their Derivatives

Using Mahler’s transcendence method, results on the algebraic independence over \(\mathbb{Q}\) of the numbers A q (z) are proved at algebraic points α with 0 < | α | < 1.

Asymptotic behavior of solutions of Poincaré difference equations

It is shown that if the zeros λ 1 , λ 2 ,..., λ n of the polynomial q(λ) = λ n + a 1 λ n−1 +... + a n are distinct and r is an integer in {1, 2,..., n} such that |λ s | ¬= |λ r | if s ¬= r, then the

Algebraic independence of the generating functions of Stern’s sequence and of its twist

Very recently, the generating function A(z) of the Stern sequence (an)n≥0, defined by a0 := 0, a1 := 1, and a2n := an, a2n+1 := an + an+1 for any integer n > 0, has been considered from the

Non-converging continued fractions related to the Stern diatomic sequence

denoted subsequently as usual by [a, a2, a4, . . . , a2 n , . . .], is transcendental. This is a consequence of Roth’s theorem and follows directly from a result of Davenport and Roth [5] concerning

A problem around Mahler functions

Let $K$ be a field of characteristic zero and $k$ and $l$ be two multiplicatively independent positive integers. We prove the following result that was conjectured by Loxton and van der Poorten


In this same journal, Coons published recently a paper [The transcendence of series related to Stern's diatomic sequence, Int. J. Number Theory6 (2010) 211–217] on the function theoretical

New approach in Mahler's method.

One of the most important motivations to write this paper originates the paper [3] by Loxton and van der Poorten in which they study algebraic independence of the values of Mahler functions. The

The Calculus of Finite Differences

  • E. N.
  • Mathematics
  • 1934
THE last edition of Boole's “Finite Differences” appeared in 1880, and was in fact a reprint of the edition of 1872. The interval of sixty years has seen in the elementary field Sheppard's


(1.1) y(n + 1) = Jy(n) + f(n, y(n)), where y is a d-vector, J is a constant d x d matrix and f(n, y) is a vector-valued function which is continuous in y for fixed n and becomes "small" in some sense


— Let K be a field of characteristic zero and k and l be two multiplicatively independent positive integers. We prove the following result that was conjectured by Loxton and van der Poorten during