On the $S_n$-invariant F-conjecture

TL;DR

This paper reduces the $S_n$-invariant F-conjecture to a polyhedral feasibility problem and uses computational methods to verify semi-ampleness and ampleness properties of divisors on moduli spaces for small n.

Contribution

It translates the conjecture into a geometric feasibility problem and provides computational evidence for divisors' semi-ampleness and ampleness on moduli spaces for n ≤ 19.

Findings

Verified semi-ampleness of $S_n$-invariant F-nef divisors for n ≤ 19

Proved that for n ≤ 16, twice any nef divisor is base-point-free

Determined the nef and semi-ample cones for specific moduli spaces

Abstract

By using classical invariant theory, we reduce the $S_{n}$-invariant F-conjecture to a feasibility problem in polyhedral geometry. We show by computer that for $n \le 19$, every integral $S_{n}$-invariant F-nef divisor on the moduli space of genus zero stable pointed curves is semi-ample, over arbitrary characteristic. Furthermore, for $n \le 16$, we show that for every integral $S_{n}$-invariant nef (resp. ample) divisor $D$ on the moduli space, $2D$ is base-point-free (resp. very ample). As applications, we obtain the nef cone of the moduli space of stable curves without marked points, and the semi-ample cone that of the moduli space of genus 0 stable maps to Grassmannian for small numerical values.