Likelihood Geometry of Reflexive Polytopes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Améndola, Carlos, Oldekop, Janike
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913440874889216
author Améndola, Carlos
Oldekop, Janike
author_facet Améndola, Carlos
Oldekop, Janike
contents We study the problem of maximum likelihood (ML) estimation for statistical models defined by reflexive polytopes. Our focus is on the maximum likelihood degree of these models as an algebraic measure of complexity of the corresponding optimization problem. We compute the ML degrees of all 4319 classes of three-dimensional reflexive polytopes, and observe some surprising behavior in terms of the presence of gaps between ML degrees and degrees of the associated toric varieties. We interpret these drops in the context of discriminants and prove formulas for the ML degree for families of reflexive polytopes, including the hypercube and its dual, the cross polytope, in arbitrary dimension. In particular, we determine a family of embeddings for the $d$-cube that implies ML degree one. Finally, we discuss generalized constructions of families of reflexive polytopes in terms of their ML degrees.
format Preprint
id arxiv_https___arxiv_org_abs_2311_13572
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Likelihood Geometry of Reflexive Polytopes
Améndola, Carlos
Oldekop, Janike
Statistics Theory
Algebraic Geometry
Combinatorics
62R01, 62F10, 52B12, 52B20, 14M25
We study the problem of maximum likelihood (ML) estimation for statistical models defined by reflexive polytopes. Our focus is on the maximum likelihood degree of these models as an algebraic measure of complexity of the corresponding optimization problem. We compute the ML degrees of all 4319 classes of three-dimensional reflexive polytopes, and observe some surprising behavior in terms of the presence of gaps between ML degrees and degrees of the associated toric varieties. We interpret these drops in the context of discriminants and prove formulas for the ML degree for families of reflexive polytopes, including the hypercube and its dual, the cross polytope, in arbitrary dimension. In particular, we determine a family of embeddings for the $d$-cube that implies ML degree one. Finally, we discuss generalized constructions of families of reflexive polytopes in terms of their ML degrees.
title Likelihood Geometry of Reflexive Polytopes
topic Statistics Theory
Algebraic Geometry
Combinatorics
62R01, 62F10, 52B12, 52B20, 14M25
url https://arxiv.org/abs/2311.13572