Primitive-root determinant densities: extremal order, dimension-zero, Fourier decay, and a lattice-smoothing no-go

Abstract

c(p): limiting fraction of n× n matrices over Fp with primitive-root determinant; 1/c(p): the rejection-sampling loss in PQ VSS; c(p)=φ(p-1)p-1Πj1(1-p-j). Over primes, c(p) follows φ(p-1)/(p-1): continuous on [0,1/2], X=12Π3\,prime(1-1/)B, B independent, (B=1)=1/(-1). Hence ∈fp c(p)=0, p xc(p)1/ x, p 1/(c(p) p)=eγ on a primorial progression. μG is singular with HμG=0, sharpening Erdős (1939) for φ(n)/n. Also 1-G(12-ε) S2 e-γ/(1/ε), S2 the twin-prime singular series (infinitude not needed), and μf=μG is Rajchman, with an explicit Fourier-decay rate under a Graham-Kolesnik envelope. E[X-1]≈2.83 vs. worst case (1+o(1))eγ p; a deterministic poly( p)-time, factoring-free two-sided certificate for 1/c(p) of gap 1+o(1); a Las Vegas generator of NTT-friendly primes q12N; and 1/c(p)>2 for all p, 1/c(p)(eγ+o(1)) y on y-friable shifts. The Micciancio-Regev smoothing ηε(Λ) has kissing floor F=(K/ε)/π/λ1(Λ*) (K the dual kissing number): at ε=2-cn every lattice has Fηεπ/((c,1)2)\,F, while at fixed ε the ratio ηε/F can diverge as n. For cyclotomic Q(ζm) ((Ring-)LWE), an exact three-case law gives dual shell gap gm∈\3/2,2,3\, infimum 3/2 on \m:ωodd(m)2\, discharging the gap hypothesis unconditionally; at ε=2-cφ(m), c>22(1+6) pins ηε(Λm) to the floor within 1+O(1/(cφ(m))).

0

Turn this paper into a full lesson

ArcXiv compiles a staged curriculum from this paper: 8-12 lessons across beginner → advanced, synthesised section guides, visuals, flashcards, a quiz, exercises, and on-demand deep dives per section. Grounded in the abstract, never invented.

Discussion (0)

Sign in to join the discussion.

Loading comments…