Neural and numerical methods for G 2 \mathrm{G_{2}} -structures on contact Calabi–Yau 7 7 –manifolds
Elli Heyes, Edward Hirst, Henrique N. Sá Earp, Tomás S. R. Silva · 2026
Read Neural and numerical methods for G 2 \mathrm{G_{2}} -structures on contact Calabi–Yau 7 7 –manifolds by Elli Heyes, Edward Hirst, Henrique N. Sá Earp, Tomás S. R. Silva (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-12438.txt.
Description
Calabi-Yau manifolds carry special geometric structures that are hard to compute, yet they matter for string theory and geometry. Elli Heyes led this work with three co-authors, presenting a numerical framework for approximating G2-structure three-forms on contact Calabi-Yau manifolds. The approach runs in three stages, moving from learned metrics to learned forms.
First, existing neural network models compute an approximate Ricci-flat metric on a Calabi-Yau threefold. Second, using that metric and an explicit construction of a G2-structure on the associated seven-dimensional Calabi-Yau link in the nine-sphere, numerical approximations of the three-form are generated at many sampled points. Finally, a dedicated neural architecture learns the three-form and its induced Riemannian metric directly from data, with torsion checked through a numerical exterior derivative.
Catalog record
| CREATOR | Elli Heyes, Edward Hirst, Henrique N. Sá Earp, Tomás S. R. Silva |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 48 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-12438