Canonical differential equations for Feynman integrals from 𝒜-hypergeometric systems in the Schwinger representation
Abstract
We follow a D-module approach to the construction of canonical differential equations for Feynman integrals directly from their Gel'fand-Kapranov-Zelevinsky (GKZ) hypergeometric systems. Starting from the Schwinger representation, we identify the associated generalized Euler integral and its GKZ system, as an alternative to the Lee-Pomeransky representation. The generalized Euler formulation allows reductions associated with facets of the Newton polytope to be identified directly, reducing the number of variables in the resulting differential systems. We construct the associated Pfaffian systems using Frobenius bases, which also provide direct access to the singular loci. After a suitable rationalizing change of variables, the systems are brought into canonical ε-form. A key feature of the construction is that the dimension of the canonical basis is determined by the holonomic rank of the GKZ system and, for the examples considered, it is smaller than or equal to the number of master integrals obtained from integration-by-parts, thus providing a reduced description of the differential system.
Related
- Differential equations for Dotsenko--Fateev integrals: the case of degenerate fields Φn,1
- Liouvillian first integrals of differential equations: A Galoisian approach
- Boundary value problems for linear differential-algebraic equations: solvability via the Kronecker canonical form
- Analysis of Delay Differential Equations Using the Offset Linear Canonical Transform
- An exact hypergeometric Weierstrass representation for Smyth's free-boundary minimal disk in the regular tetrahedron
BibTeX
@misc{jimenezsantacruz2026canonical,
title = {{Canonical differential equations for Feynman integrals from $\mathcalA$-hypergeometric systems in the Schwinger representation}},
author = {Mateo Jimenez-Santacruz and Cristhiam Lopez-Arcos and Alexander Quintero Velez},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.16107},
eprint = {2609.16107},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.16107},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Canonical differential equations for Feynman integrals from 𝒜-hypergeometric systems in the Schwinger representation
AU - Mateo Jimenez-Santacruz
AU - Cristhiam Lopez-Arcos
AU - Alexander Quintero Velez
PY - 2026
DA - 2026/09/14
PB - arXiv
DO - 10.48550/arxiv.2609.16107
UR - https://arxiv.org/abs/2609.16107
AB - We follow a D-module approach to the construction of canonical differential equations for Feynman integrals directly from their Gel'fand-Kapranov-Zelevinsky (GKZ) hypergeometric systems. Starting from the Schwinger representation, we identify the associated generalized Euler integral and its GKZ system, as an alternative to the Lee-Pomeransky representation. The generalized Euler formulation allows reductions associated with facets of the Newton polytope to be identified directly, reducing the number of variables in the resulting differential systems. We construct the associated Pfaffian systems using Frobenius bases, which also provide direct access to the singular loci. After a suitable rationalizing change of variables, the systems are brought into canonical ε-form. A key feature of the construction is that the dimension of the canonical basis is determined by the holonomic rank of the GKZ system and, for the examples considered, it is smaller than or equal to the number of master integrals obtained from integration-by-parts, thus providing a reduced description of the differential system.
ER -