Machine Learning Calabi-Yau Volumes

Abstract: The talk will give an overview of our work from 2017 which introduced machine learning techniques in string theory. This work made use of standard machine learning techniques, including convolutional neural networks (CNN), in order to find new formulas for the minimum volume of Sasaki-Einstein manifolds corresponding to toric Calabi-Yau 3-folds. These geometries, by the AdS/CFT correspondence, relate to a large class of $4d N=1$ supersymmetric gauge theories. The R-charges of the dual gauge theories are known to be related to the minimum volumes of the corresponding Sasaki-Einstein manifolds. In this talk, we will review the process of volume minimization and give a short overview on ongoing work.

condensed matterHEP - theory

Audience: researchers in the topic

Numerical Methods in Theoretical Physics