| Speaker: | Damian Kaloni Mayorga Pena (Instituto Superior Tecnico, Univ. of Lisbon) |
|---|---|
| Title: | Machine learning for heterotic string phenomenology |
| Date (JST): | Tue, Dec 17, 2024, 13:30 - 15:00 |
| Place: | Seminar Room A |
| Abstract: | In this talk I will present recent advancements in applying machine learning to Calabi-Yau (CY) compactifications of the E8 × E8 heterotic string. A central challenge in this domain is the determination of the Ricci-flat metric on the CY, which remains analytically elusive. I will discuss various approaches that have been developed to approximate these metrics numerically, framing them as optimization problems solvable by machine learning techniques. Additionally, I will showcase numerical methods for constructing globally defined harmonic one-forms on Calabi-Yau manifolds and their relevance in computing kinetically normalized Yukawa couplings. I will discuss the efficacy and limitations of these techniques as well as the physical results in concrete examples such as the Tian-Yau manifold. |
