Algebraic aspects of the polynomial Littlewood-Offord problem

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Jin, Zhihan, Kwan, Matthew, Sauermann, Lisa, Wang, Yiting
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918038052274176
author Jin, Zhihan
Kwan, Matthew
Sauermann, Lisa
Wang, Yiting
author_facet Jin, Zhihan
Kwan, Matthew
Sauermann, Lisa
Wang, Yiting
contents Consider a degree-$d$ polynomial $f(ξ_1,\dots,ξ_n)$ of independent Rademacher random variables $ξ_1,\dots,ξ_n$. To what extent can $f(ξ_1,\dots,ξ_n)$ concentrate on a single point? This is the so-called polynomial Littlewood-Offord problem. A nearly optimal bound was proved by Meka, Nguyen and Vu: the point probabilities are always at most about $1/\sqrt n$, unless $f$ is "close to the zero polynomial" (having only $o(n^d)$ nonzero coefficients). In this paper we prove several results supporting the general philosophy that the Meka-Nguyen-Vu bound can be significantly improved unless $f$ is "close to a polynomial with special algebraic structure", drawing some comparisons to phenomena in analytic number theory. In particular, one of our results is a corrected version of a conjecture of Costello on multilinear forms (in an appendix with Ashwin Sah and Mehtaab Sawhney, we disprove Costello's original conjecture).
format Preprint
id arxiv_https___arxiv_org_abs_2505_23335
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Algebraic aspects of the polynomial Littlewood-Offord problem
Jin, Zhihan
Kwan, Matthew
Sauermann, Lisa
Wang, Yiting
Combinatorics
Number Theory
Probability
Consider a degree-$d$ polynomial $f(ξ_1,\dots,ξ_n)$ of independent Rademacher random variables $ξ_1,\dots,ξ_n$. To what extent can $f(ξ_1,\dots,ξ_n)$ concentrate on a single point? This is the so-called polynomial Littlewood-Offord problem. A nearly optimal bound was proved by Meka, Nguyen and Vu: the point probabilities are always at most about $1/\sqrt n$, unless $f$ is "close to the zero polynomial" (having only $o(n^d)$ nonzero coefficients). In this paper we prove several results supporting the general philosophy that the Meka-Nguyen-Vu bound can be significantly improved unless $f$ is "close to a polynomial with special algebraic structure", drawing some comparisons to phenomena in analytic number theory. In particular, one of our results is a corrected version of a conjecture of Costello on multilinear forms (in an appendix with Ashwin Sah and Mehtaab Sawhney, we disprove Costello's original conjecture).
title Algebraic aspects of the polynomial Littlewood-Offord problem
topic Combinatorics
Number Theory
Probability
url https://arxiv.org/abs/2505.23335