An inverse theorem for Freiman multi-homomorphisms

arXiv:2002.11667 · math.CO · Submitted 2020-02-26 · Read on arXiv

math.CO

Submitted: 2020-02-26

Updated: 2026-11-01

Comments: 199 pages, significantly improved exposition and expanded the motivation of the arguments

License: http://creativecommons.org/licenses/by/4.0/

The gist: Let G 1,, G k and H be vector spaces over a finite field F p of prime order.

Terminology

Abstract

Let G 1,, G k and H be vector spaces over a finite field F p of prime order. Let A G 1 times times G k be a set of size δG 1 G k. Let a map ϕ A to H be a multi-homomorphism, meaning that for each direction d in [k], and each element (x 1,, x d-1, x d+1,, x k) of G 1 times times G d-1 times G d+1 times times G k, the map that sends each y d such that (x 1,, x d-1, y d, x d+1,, x k) in A to ϕ(x 1,, x d-1, y d, x d+1,, x k) is a Freiman homomorphism (of order 2). In this paper, we prove that for each such map, there is a multiaffine map Φ G 1 times times G k to H such that ϕ= Φ on a set of density ((O k(1))(O k,p(δ-1)))-1, where(t) denotes the t-fold exponential. Applications of this theorem include: a quantitative inverse theorem for approximate polynomials mapping G to H, for finite-dimensional F p-vector spaces G and H, in the high-characteristic case, a quantitative inverse theorem for uniformity norms over finite fields in the high-characteristic case, and a quantitative structure theorem for dense subsets of G 1 times times G k that are subspaces in the principal directions (without additional characteristic assumptions).

Sources

Related papers