Burgess bounds for short character sums evaluated at forms

From MaRDI portal
Publication:2201817



Abstract: In this work we establish a Burgess bound for short multiplicative character sums in arbitrary dimensions, in which the character is evaluated at a homogeneous form that belongs to a very general class of "admissible" forms. This n-dimensional Burgess bound is nontrivial for sums over boxes of sidelength at least , with . This is the first Burgess bound that applies in all dimensions to generic forms of arbitrary degree. Our approach capitalizes on a recent stratification result for complete multiplicative character sums evaluated at rational functions, due to the second author.


In [Proc. Lond. Math. Soc. 12, 193--206 (1962; Zbl 0106.04004)], using Hasse's result on an \(L\)-function of an algebraic function field and Weil's bound resulting from his proof of the Riemann hypothesis for algebraic curve over finite fields, Burgess established his famous bound for short Dirichlet character sums which is better than the Pólya-Vinogradov inequality when the length of the interval is smaller than the square-root of the modulus. The paper under review deals with a multidimensional Burgess bound short character sums evaluated at generic forms of arbitrary degrees. More precisely, let \(d \in \mathbb{Z}_{\geqslant 1}\), \(\chi\) be a non-principal Dirichlet character modulo a prime \(p\) of order \(\Delta\), and \(F \in \mathbb{Z}[n_1,\dotsc,n_d]\) be a \(d\)-dimensional form of degree \(D\) satisfying the following condition: if the reduction \(\overline{F}\) modulo \(p\) of \(F\) is written as \(\overline{F} = G^\Delta H\) with \(G,H \in \mathbb{F}_p[n_1,\dotsc,n_d]\) such that \(H\) is \(\Delta\)-power-free over \(\mathbb{F}_p\), then there is no \(A \in \textrm{GL}(n,\mathbb{F}_p)\) such that \(H(\mathbf{n}A) \in \mathbb{F}_p[n_1,\dotsc,n_d]\). For all \(r \in \mathbb{Z}_{\geqslant 1}\), set \(\theta_r := r\) if \(d=1\), \(\theta_r := \left \lfloor \frac{r-1}{d-1} \right \rfloor\) if \(d \geqslant 2\). Let \((y_1,\dotsc,y_d) \in \mathbb{R}_{\geqslant 1}^d\) such that \(y_{\min} y_{\max} < p^{1 + \frac{1}{2 \theta_r}}\). Then, uniformly for all \((x_1,\dotsc,x_d) \in \mathbb{R}_{\geqslant 1}^d\) and all \(r \in \mathbb{Z}_{\geqslant 1}\), we have \[\sum_{x_1 < n_1 \leqslant x_1+y_1} \dotsb \sum_{x_d < n_d \leqslant x_d+y_d} \chi \left( F(n_1,\dotsc,n_d)\right) \ll \left( y_1 \dotsb y_d \right)^{1-\frac{1}{2r}} \ y_{\min}^{- \frac{1}{2r}} \ p^{\frac{d \theta_r+1}{4r \theta_r}} \ (\log p)^{d+1}\] where the implied constant depends only on \(D\), \(d\), \(\Delta\) and \(r\). The proof essentially follows Burgess' ideas but, as pointed out by the authors, the generalization of these ideas to the \(d\)-dimensional setting makes a highly singular projective variety appear to which Weil bounds cannot be applied. It is then explained how the second author's recent work [Int. Math. Res. Not. 10, 2881--2917 (2020; Zbl 1477.11153)] is suitable in this situation and allows them to overcome this problem.



Cites work









This page was built for publication: Burgess bounds for short character sums evaluated at forms

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2201817)