K-theoretic positivity for matroids

Christopher Eur,Matt Larson

2024-09-19

Abstract:Hilbert polynomials have positivity properties under favorable conditions. We establish a similar "K-theoretic positivity" for matroids. As an application, for a multiplicity-free subvariety of a product of projective spaces such that the projection onto one of the factors has birational image, we show that a transformation of its K-polynomial is Lorentzian. This partially answers a conjecture of Castillo, Cid-Ruiz, Mohammadi, and Montano. As another application, we show that the h*-vector of a simplicially positive divisor on a matroid is a Macaulay vector, affirmatively answering a question of Speyer for a new infinite family of matroids.

Algebraic Geometry,Combinatorics

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is to establish a similar "K - theoretic positivity" in matroids. Specifically, the author aims to prove that for a multiplicity - free subvariety satisfying certain conditions, the transformation of its K - polynomial is Lorentzian. In addition, the article also explores the question of whether the \(h^*\)-vector of a simplicially positive divisor on a matroid is a Macaulay vector.

### Main Contributions

1. **K - theoretic Positivity**:
- The author proves that "K - theoretic positivity", which is similar to the positivity property of Hilbert polynomials under favorable conditions, also holds for matroids.
- When specifically applied to multiplicity - free subvarieties, if the mapping projected onto a factor is generically finite, then the transformation of its K - polynomial is Lorentzian.

2. **Application of Lorentzian Property**:
- By proving that the Snapper polynomial is Lorentzian after a specific transformation, it partially answers the conjecture of Castillo, Cid - Ruiz, Mohammadi and Montaño.

3. **Study of \(h^*\)-vector**:
- For simplicially positive divisors, the author proves that its \(h^*\)-vector is a Macaulay vector, thus positively answering a question about a new infinite family of matroids proposed by Speyer.

### Summary of Mathematical Formulas

- **Relationship between Hilbert Series and \(h^*\)-vector**:
\[
\sum_{k \geq 0}|\text{lattice points in }kQ|q^k=\frac{h^*_0(Q)+h^*_1(Q)q+\cdots +h^*_d(Q)q^d}{(1 - q)^{d + 1}}
\]

- **Euler Characteristic Value Calculation**:
\[
\chi(M,L^{\otimes t_1}_{S_1}\otimes\cdots\otimes L^{\otimes t_m}_{S_m})=\sum_k(-1)^{r-|k|}a_kt[k]
\]

- **Definition of Lorentzian Polynomial**:
\[
f=\sum_kc_kt^k\in\mathbb{R}[t_1,\ldots,t_m]
\]
is Lorentzian if:
1. The support set \(\{k\in\mathbb{Z}_{\geq 0}^m:c_k>0\}\) is equal to the base polytope \(B(P)\cap\mathbb{Z}^m\) of some matroid \(P\).
2. Any \((d - 2)\)-th order partial derivative is a quadratic form with at most one positive eigenvalue.

- **Normalization Formula**:
\[
N(f)=\sum_k\frac{c_kt^k}{k!}
\]

Through these results, the author not only promotes the study of K - theoretic positivity in matroid theory, but also provides new tools and perspectives for understanding the geometric and combinatorial properties of matroids.