Solution-count conjecture for Vinogradov systems modulo NN

Let γ(t)=(t,t2,,tn)\gamma(t)=(t,t^2,\ldots,t^n) and define Φ(t)=i=1nγ(ti)\Phi(\vec t)=\sum_{i=1}^n\gamma(t_i). For t[Z/NZ]n\vec t\in[\mathbb Z/N\mathbb Z]^n, let N(t;N)\mathbf N(\vec t;N) be the number of s[Z/NZ]n\vec s\in[\mathbb Z/N\mathbb Z]^n satisfying Φ(s)=Φ(t)\Phi(\vec s)=\Phi(\vec t). Use tjtk|t_j-t_k| for the least nonnegative representative of the difference's absolute value in the source's ring metric. Solution-count conjecture for Vinogradov systems modulo NN. For all ε>0\varepsilon>0 there exists a constant Cε,n>0C_{\varepsilon,n}>0 such that

N(t;N)min{Cε,nNε+n(n1)/21j<kntjtk1,Nn}.\mathbf N(\vec t;N)\leq\min\left\{C_{\varepsilon,n}N^{\varepsilon+n(n-1)/2}\prod_{1\leq j<k\leq n}|t_j-t_k|^{-1},\,N^n\right\}.

This conjecture is suggested by comparison with the Euclidean Jacobian estimate and would control the multiplicities in the discrete convolution argument; the source does not establish it.

Sources & referencesView supporting material

Primary source

Jonathan Hickman and James Wright, “The Fourier restriction and Kakeya problems over rings of integers modulo N”, arXiv:1801.03176 (2018).

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.