2
$\begingroup$

Let $\alpha,\beta\in\overline{\mathbb Q}$ be such that $0<|a|<1$ and $|\beta|=1$. One assumes that $\beta$ is not a root of unity. Is $A:=\sum_{n\ge0}\frac{\alpha^{2^n}}{2-\beta^{2^n}}$ transcendental?

If one introduces the function $f(X,Y)=\sum_{n\ge0}\frac{X^{2^n}}{2-Y^{2^n}}$, one sees that $f(X^2,Y^2)=f(X,Y)-\frac{X}{2-Y}$, $f$ is analytic on $\{(X,Y)\in\mathbb C^2\mid |X|<1\text{ and }|Y|<1\}$ and is definite on $\{(X,Y)\in\mathbb C^2\mid |X|<1\text{ and }|Y|=1\}$

But one can not use Mahler's theorem (see Nishioka's book "Mahler functions and transcendence", Chpater II) to obtain the transcendence of $A$ since $|\beta|\ge1$.

Does anyone know a way to prove $A$ is transcendental?

$\endgroup$
1
  • $\begingroup$ All proofs of transcendence we know of follows these lines: assume to the contrary that $A$ is algebraic, and let $g(x)$ be its minimal polynomial. Does this assumption eventually lead to the existence of a rational integer $k$ satisfying $0 < k < 1$, which obviously does not exist? If so you get a contradiction which shows $A$ is transcendental. $\endgroup$ Jul 17, 2020 at 12:08

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.