Soprunova's mean-value formula conjecture for exponential sums

About 22 years old · traced to

Let Λ\Lambda be a finite subset of Rn\mathbb R^n, and let F1,…,FnF_1,\dots,F_n be exponential sums with real frequencies whose Newton polyhedra Δ1,…,Δn\Delta_1,\dots,\Delta_n form a developed collection. Let GG be another exponential sum with real frequencies. If MΩM_{\Omega} denotes the mean value of GG over the zeros of F1=⋯=Fn=0F_1=\cdots=F_n=0, let Δ=Δ1+⋯+Δn\Delta=\Delta_1+\cdots+\Delta_n, and for each vertex α\alpha of Δ\Delta let CαC_\alpha be the constant term defined from the formal exponential series associated with F1⋯FnF_1\cdots F_n and GG. Let kαk_\alpha be the corresponding combinatorial coefficients.

Mean-value formula conjecture. The mean value is equal to

MΩ=1(−2π)n∑αkαCα,M_{\Omega}=\frac{1}{(-2\pi)^n}\sum_{\alpha}k_\alpha C_\alpha,

where the sum is over the vertices α\alpha of Δ\Delta.

This is the real-frequency analogue of the formula known for rational frequencies. The conjecture concerns whether the explicit combinatorial and coefficient expression remains valid for the mean value over zeros of a system with arbitrary real frequencies.

References

Primary source

Evgenia Soprunova, “Zeros of systems of exponential sums and trigonometric polynomials”, arXiv:math/0401074 (2004).

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.