Projective linear configurations via non-reductive actions

TL;DR

This paper develops a new algebraic uniformization for blow-ups of projective space along linear subspaces, linking algebraic and geometric properties via non-reductive group actions, with applications to moduli spaces.

Contribution

It introduces an $A^1$-homotopy based model for these blow-ups, providing a unified coordinate system and explicit descriptions of Cox rings, extending known results and offering algorithmic tools.

Findings

Cox ring is an invariant subring of a graded polynomial ring.

Provides a geometric proof and topological intuition for Mukai's theorem.

Algorithmically describes Cox(X) and verifies finite generation when applicable.

Abstract

We study the iterated blow-up X of projective space along an arbitrary collection of linear subspaces. By replacing the universal torsor with an $\mathbb{A}^1$-homotopy equivalent model, built from $\mathbb{A}^1$-fiber bundles not just algebraic line bundles, we construct an "algebraic uniformization": X is a quotient of affine space by a solvable group action. This provides a clean dictionary, using a single coordinate system, between the algebra and geometry of hypersurfaces: effective divisors are characterized via toric and invariant-theoretic techniques. In particular, the Cox ring is an invariant subring of a Pic(X)-graded polynomial ring and it is an intersection of two explicit finitely generated rings. When all linear subspaces are points, this recovers a theorem of Mukai while also giving it a geometric proof and topological intuition. Consequently, it is algorithmic to describe Cox(X) up to any degree, and when Cox(X) is finitely generated it is algorithmic to verify finite generation and compute an explicit presentation. We consider in detail the special case of $\overline{M}_{0,n}$. Here the algebraic uniformization is defined over the integers. It admits a natural modular interpretation, and it yields a precise sense in which $\overline{M}_{0,n}$ is "one non-linearizable $\mathbb{G}_a$ away" from being a toric variety. The Hu-Keel question becomes a special case of Hilbert's 14th problem for $\mathbb{G}_a$.