Deprecated: $wgMWOAuthSharedUserIDs=false is deprecated, set $wgMWOAuthSharedUserIDs=true, $wgMWOAuthSharedUserSource='local' instead [Called from MediaWiki\HookContainer\HookContainer::run in /var/www/html/w/includes/HookContainer/HookContainer.php at line 135] in /var/www/html/w/includes/Debug/MWDebug.php on line 372
On simultaneous Diophantine approximation - MaRDI portal

On simultaneous Diophantine approximation (Q2145923)

From MaRDI portal
scientific article
Language Label Description Also known as
English
On simultaneous Diophantine approximation
scientific article

    Statements

    On simultaneous Diophantine approximation (English)
    0 references
    0 references
    15 June 2022
    0 references
    Let \(\boldsymbol{\alpha}=(\alpha_1,\ldots,\alpha_m)\in {\mathbb R}^m\) be such that \(1,\alpha_1,\ldots,\alpha_m\) are linearly independent over \({\mathbb Q}\) and for a positive integer \(x\) put \[ b(x)=\max\{ \| \alpha_1 x\|,\ldots, \|\alpha_m x\|\}. \] By Dirichlet's theorem for any positive real \(q\) there is \(x\in {\mathbb Z}_{\ge 1}\), \(x\le e^{mq}\) such that \(b(x)\le e^{-q}\). In particular, \(x b(x)^m\le 1\) holds for infinitely many positive integers \(x\). Let \(1=x_1<x_2<\cdots\) be the positive integers which are best possible for the above inequality, namely that for all \(x<x_j\) we have \(b(x)\ge b(x_{j-1})\). Call \(\boldsymbol{\alpha}\) badly approximable if \(x b(x)^m\gg 1\). Set \(b_j:=b(x_j)\) and consider the following conditions on \(\boldsymbol{\alpha}\): \begin{itemize} \item[(i)] \({\alpha}\) is badly approximable. \item[(ii)] \(x_{j+1}/x_j\ll 1.\) \item[(iii)] \(b_j/b_{j+1}\ll 1\). \end{itemize} When \(m=1\) it is known that the above conditions are equivalent and recent work of \textit{R. Akhunzhanov} and \textit{N. Moshchevitin} [Math. Z. 301, No. 2, 1573--1602 (2022; Zbl 1492.11110)] shows that for every \(m\ge 1\), (i) implies (ii) and (iii). The main result of the paper is to show that for every \(m\) there are \(\boldsymbol{\alpha}\) with (iii) but not (i) or (ii). The proof uses a theorem of \textit{D. Roy} [Ann. Math. (2) 182, No. 2, 739--786 (2015; Zbl 1328.11076)] on parametric geometry of numbers.
    0 references
    0 references
    approximation of \(m\)-tuples
    0 references
    continued fractions
    0 references
    badly approximable numbers
    0 references
    geometry of numbers
    0 references

    Identifiers