Targeting Multi-Loop Integrals with Neural Networks
High Energy Physics - Phenomenology
2023-05-22 v3
Abstract
Numerical evaluations of Feynman integrals often proceed via a deformation of the integration contour into the complex plane. While valid contours are easy to construct, the numerical precision for a multi-loop integral can depend critically on the chosen contour. We present methods to optimize this contour using a combination of optimized, global complex shifts and a normalizing flow. They can lead to a significant gain in precision.
Cite
@article{arxiv.2112.09145,
title = {Targeting Multi-Loop Integrals with Neural Networks},
author = {Ramon Winterhalder and Vitaly Magerya and Emilio Villa and Stephen P. Jones and Matthias Kerner and Anja Butter and Gudrun Heinrich and Tilman Plehn},
journal= {arXiv preprint arXiv:2112.09145},
year = {2023}
}
Comments
20 pages, 9 figures, v3: added two references