Bourgain–Chang bilinear estimate conjecture over finite fields

Let pp be a prime, let Fp\mathbb{F}_p be the finite field with pp elements, and define

T(f1,f2)(s)=ns,nFpf1(sn)f2(n)K(sn,n),T(f_1,f_2)(s)=\sum_{n\ne s,\,n\in\mathbb{F}_p}f_1(s-n)f_2(n)K(s-n,n),

where

K(a,b)=1pyFpep(ay2+by),ep(x)=e2πix/p.K(a,b)=\frac{1}{p}\sum_{y\in\mathbb{F}_p}e_p(ay^2+by),\qquad e_p(x)=e^{-2\pi i x/p}.

Bourgain–Chang bilinear estimate conjecture. For every ε>0\varepsilon>0, there exists CεC_\varepsilon such that, for all sufficiently large primes pp,

T(f1,f2)2Cεpε1/2f12f22.\|T(f_1,f_2)\|_2\leq C_\varepsilon p^{\varepsilon-1/2}\|f_1\|_2\|f_2\|_2.

This conjecture concerns the expected near-optimal L2×L2L2L^2\times L^2\to L^2 bound for a bilinear operator whose kernel is a normalized quadratic Gauss sum. The source attributes it to Bourgain and Chang; the paper improves an existing estimate but does not state that this conjecture is resolved.

Sources & referencesView supporting material

Primary source

Necef Kavrut and Shukun Wu, “A bilinear estimate in F_p”, arXiv:2401.07925 (2024).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.