Machine learning Kreuzer–Skarke Calabi–Yau threefolds
By using a fully connected feedforward neural network, we study topological invariants of a class of Calabi–Yau manifolds constructed as hypersurfaces in toric varieties associated with reflexive polytopes from the Kreuzer–Skarke database. In particular, we find the existence of a simple expression for the Euler number that can be learned in terms of limited data extracted from the polytope and its dual.
Authors
- B. Campbell (ORCID: https://orcid.org/0009-0004-2331-1802)
- Vishnu Jejjala (ORCID: https://orcid.org/0000-0003-2603-6717)
- Per Berglund (ORCID: https://orcid.org/0000-0003-1675-4133)
Institutions
- University of New Hampshire (US)
- University of the Witwatersrand (ZA)
- National Institute for Theoretical Physics (ZA)
Publication Details
- Journal
- International Journal of Modern Physics A
- Published
- 2025-08-21
- DOI
- https://doi.org/10.1142/s0217751x25500976
- Citations
- 2
- Primary Topic
- Advanced Algebra and Geometry
- Type
- article
- Field-Weighted Citation Impact
- 3.69
Funders
- U.S. Department of Energy