Waring's Problem with Polynomial Summands
Let \(f(x)\) be an integral-valued polynomial (that is, \(f(x)\) is an integer for every integer \(x\)) of the form \[ f(x)=\sum_{i=1}^t a_ix^{k_i}, \qquad k=k_1>k_2>\cdots>k_t, \eqno(1) \] and denote by \(\bar G(f)\) the least \(s\) such that the equation \[ f(x_1)+f(x_2)+\cdots+f(x_s)=n \] is solvable in positive integers \(x_1,\dots,x_s\) for every sufficiently large integer \(n\) satisfying some naturally required congruence condition (this condition on \(n\) is explicitly described in connection with the associated ``singular series). This \(\bar G(f)\) may be compared with the familiar notation \(G(k)\) in Waring's problem which is defined to be the least \(s\) such that every sufficiently large integer is written as a sum of \(s\) positive \(k\)-th powers. It may be proved by rather classical techniques that \(\bar G(f)\leq (4+o(1))k\log k\) as \(k\rightarrow\infty\). No substantial improvement on this bound is known for general polynomials \(f\), while \textit{T. D. Wooley} proved recently that \(\bar G(f)\leq (1+o(1))k\log k\) when both \(t\) and \(k_2\) in (1) are very small, roughly speaking [J. Reine Angew. Math. 488, 79-140 (1997; Zbl 0880.11060)]. The latter bound is as sharp as Wooley's upper bound for \(G(k)\). In this paper, the author establishes the bound \[ \bar G(f)\leq k(\log k+2\log t+\log(k-k_t)+\log\log k+O(1)). \] When \(\log(k-k_t)=o(\log k)\), therefore, this estimate is as strong as the best known bound for \(G(k)\). Namely he widen the class of polynomials for which the bound \(\bar G(f)\leq (1+o(1))k\log k\) holds. The first key ingredient of the proof is to modify the efficient differencing method of \textit{T. D. Wooley} [Ann. Math. 135, 131-164 (1992; Zbl 0754.11026)] to handle general polynomials, which yields mean value estimates of the same strength as in the case \(f(x)=x^{k_t}\). The second one is to combine the latter result with the diminishing ranges method of Hardy and Littlewood.
- Waring's problem for polynomials of small degree
- Greedy algorithm with gaps
- Additive properties for sets of polynomial values
- On Waring's problem with polynomial summands
- Weyl sums, mean value estimates, and Waring's problem with friable numbers
- EQUAL SUMS OF LIKE POLYNOMIALS
- Waring's problem in function fields
- Waring's problem with shifts
- On Waring's problem with polynomial summands II
- The application of a new mean value theorem to the fractional parts of polynomials
- scientific article; zbMATH DE number 3997899 (Why is no real title available?)
- scientific article; zbMATH DE number 817487 (Why is no real title available?)
- scientific article; zbMATH DE number 3025867 (Why is no real title available?)
- Additive bases arising from functions in a Hardy field
- Solutions to polynomial congruences with variables restricted to a box
This page was built for publication: Waring's Problem with Polynomial Summands
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4520425)
