scientific article; zbMATH DE number 3141382

From MaRDI portal
Publication:3259124





Let \(n\), \(a_0, a_1, \ldots, a_n\) be integers such that \(n > 0\), \(H = \max \vert a_0\vert, \vert a_1\vert, \ldots, \vert a_n\vert) > 0\). The author states that, by improving an earlier method of his [Dokl. Akad. Nauk SSSR, n. Ser. 66, 565--567 (1949; Zbl 0038.02903); Am. Math. Soc., Transl., II. Ser. 59, 224--245 (1966); translation from Izv. Akad. Nauk SSSR, Ser. Mat. 15, 53--74 (1951; Zbl 0042.04802)] he can prove: If \(\alpha\ne 0, \ne1\), is an algebraic number, then, for \(n^4 < \log H\), \[ \vert a_0+a_1\log \alpha + \ldots + a_n(\log \alpha)^n\vert > H^{-\gamma_0 n^2(\log(n+1))^3}. \] Similarly, under the same relation between \(n\) and \(H\), \[ \vert a_0+a_1\pi + \ldots + a_n\pi^n\vert > H^{-\gamma n\log(n+1)}. \] Here \(\gamma_0 > 0\) depends only on \(\alpha\), and \(\gamma > 0\) is an absolute constant. Both inequalities greatly improve the so far best results due to \textit{K. Mahler} [Philos. Trans. R. Soc. Lond., Ser. A 245, 371--398 (1953; Zbl 0052.04404)].











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3259124)