Polynomial-rank symmetrization conjecture for multilinear forms

At least 4 years old · documented by

Let VV be a vector space over Fp\mathbb F_p, let kk be a positive integer, and let T ⁣:Vk→FpT\colon V^k\to\mathbb F_p be a kk-linear form. For each permutation π∈Sk\pi\in\mathfrak S_k, write TπT_\pi for the form obtained by permuting the arguments of TT, and let prank⁡\operatorname{prank} denote partition rank. A form is symmetric if it is invariant under all permutations of its arguments.

Symmetrization conjecture. If

prank⁡(T−Tπ)≤r\operatorname{prank}(T-T_\pi)\leq r

for every π∈Sk\pi\in\mathfrak S_k, then there exists a symmetric kk-linear form S ⁣:Vk→FpS\colon V^k\to\mathbb F_p such that

prank⁡(S−T)≤r′,\operatorname{prank}(S-T)\leq r',

where r′r' is a polynomial function of rr.

The conjecture concerns quantitative stability of symmetry under partition rank. It is only interesting when p≤kp\leq k, since for p>kp>k averaging over permutations gives a symmetric form with the required bound. Combined with work of Gowers and Milićević, it would provide the missing ingredient for quantitative bounds in the Up+1U^{p+1}-inverse theorem; the source presents it as surprisingly difficult and potentially independently interesting.

References

Primary source

Jonathan Tidor, “Quantitative bounds for the U^4-inverse theorem over low characteristic finite fields”, arXiv:2109.13108 (2022).

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.